Week 15 DVOA Ratings

Photo: USA Today Sports Images
So, let's talk about Green Bay some more. Their narrow win over the Giants actually drops their overall season DVOA slightly. I'll admit, the whole Packers situation is a bit frustrating. The majority of the football commentariat describes Green Bay as unbeatable, as head and shoulders ahead of the rest of the league. Those of us who look at things a little closer know that the Packers have flaws. We know there is a difference between "unbeatable" and "unbeaten." And yet... let's face it, it just looks weird to not have an undefeated 12-0 team on top of the DVOA ratings. It's even worse now that they've dropped to third in Weighted DVOA. Doesn't Green Bay look like the best team in the league to all of us, even if we don't think they should be ranked among the best teams of all time?

That's a quote from my DVOA commentary after Week 13 of the 2011 season. Does it sound familiar?

I've detailed my frustrations about the 2020 Kansas City Chiefs, both in Audibles each week and in the DVOA commentaries. The majority of the football commentariat believes the Kansas City Chiefs are massive favorites to win the Super Bowl, that there's a chasm between the Chiefs and the rest of the league. "If it's Chiefs vs. The Field to win the Super Bowl, most would take the Chiefs" according to Pro Football Talk this week. "The Chiefs are inevitable and they just feel unbeatable," wrote our old friend Charles McDonald this week. Even analytics people on Twitter maintain that the Chiefs are dominant. "I just don't see how any team except maybe -- MAYBE -- the Packers can beat this Chiefs team," said Ben Baldwin on Twitter Sunday.

And I'm frustrated and feel like I'm going a little bit crazy because every week the Chiefs play a close game and they don't pull away from the rest of the league in our DVOA ratings. It's not just that the Chiefs aren't way ahead of everyone else; the Chiefs aren't even No. 1. And just like the 2011 Packers after Week 13... the Chiefs had a narrow win this week that actually dropped their overall season DVOA slightly. Yes, that's right, the Saints ended up with a higher DVOA on Sunday than the Chiefs did. It was 39.8% total for the Saints, 12.9% for the Chiefs. The Saints outgained the Chiefs 5.5 yards per play to 4.7, while the Chiefs had the higher success rate, 49% to 40%. Add in the Chiefs' fumbles and the Drew Brees interception, and the offense and defense ratings were basically even. The difference comes in the special teams. I did not count the free kick at the end of the first half, or else the gap would have been even larger. Losing seven yards on a punt return is bad, and getting the ball stripped is worse.

Do you feel like that overrates special teams? Maybe it does, but that doesn't matter to Kansas City's overall DVOA. The Chiefs rank third in total DVOA behind New Orleans and Tampa Bay, and second in weighted DVOA behind New Orleans. If we took out special teams, they would rank... exactly the same, because right now Kansas City has a special teams DVOA of 0.0%.

Trying to figure out why my opinion of the Chiefs is so much different from everybody else's opinion of the Chiefs, I brainstormed for similar teams and came up with the 2011 Packers. But I didn't realize quite how similar these two teams were until I went back and looked at what I wrote about the Packers while the 2011 season was going on.

In order to talk about Green Bay properly, I think we need to separate the concepts of "greatness" and "dominance." The Packers are by no means a dominant team. They are fairly one-dimensional; that dimension just happens to be performing off the charts... Some readers have suggested that perhaps total DVOA is not the proper measure to use when a team is so superlative in one area of the game. That's the "defense doesn't matter, because Aaron Rodgers can outscore anybody" theory.

Now this should really be sounding familiar.

Aaron Rodgers spent the 2011 season outscoring everybody... until he suddenly didn't. First came Week 15, when the Packers lost their first game of the season. Ironically, the team they lost to was Kansas City. The Chiefs sacked Rodgers four times and kept him to 235 passing yards while Kyle Orton -- remember when he played four games for Kansas City? -- had 299 passing yards and the Chiefs had four field goals to win 19-14.

It didn't really matter, though. That was considered a blip by pretty much everyone. The Packers were still prohibitive favorites to win the Super Bowl going into the 2011 playoffs. They were 8-point favorites over the Giants in their first playoff game. Then they lost 37-20. Eli Manning torched the Packers defense for 330 passing yards and three touchdowns. The Packers fumbled the ball three times and the Giants recovered all three. And just like that, the Green Bay Packers went home, and their attempt to defend the title was over.

If the Packers are not a dominant team, how have they managed to go 12-0 this season? The main answer is consistency. The Packers have been absurdly consistent. After this week, the Packers lead the league with 3.3% variance.

It was just last week that I wrote about the fact that Kansas City has been the most consistent team in the NFL this year from week to week. After the Saints game, they now lead the NFL with a 3.9% variance. It's just one of the many similarities between the 2020 Chiefs and the 2011 Packers. Both teams were the defending champions, so they "took every team's best blow." Both teams were led by all-time great quarterbacks having all-time great passing seasons. Both teams were mediocre running the ball and defending against the pass. Both teams were porous defending against the run. This table looks at the current version of DVOA and has numbers through Week 15, matching this year's Chiefs with the Packers after their only regular-season loss, By that point, Green Bay was on top of the DVOA ratings, but just barely.

2011 GB vs. 2020 KC after Week 15
Year 2011 2020
Team GB KC
W-L 13-1 13-1
Total DVOA 24.7% (1) 23.4% (3)
Pass Offense 61.7% (1) 53.4% (1)
Run Offense 0.7% (10) -5.5% (13)
Pass Defense 9.2% (20) 3.1% (13)
Run Defense 5.5% (29) 2.6% (30)
Special Teams 2.2% (7) 0.0% (17)

You know how the Chiefs make games close by giving up points late when they already have a very high win probability? That's one of the arguments as to why the Chiefs are more dominant than their close wins would suggest. Turns out the 2011 Packers did the same thing. In Week 2, they came back from a 13-0 deficit but the game ended close, a 30-23 Packers win, after a Cam Newton rushing touchdown with 37 seconds left. In Week 7, they beat the Vikings by only six points, 33-27, because Minnesota scored 10 unanswered points in the fourth quarter. In Week 9, a 45-24 lead ended up a 45-38 victory after Philip Rivers hit Vincent Jackson twice in a 1:07 span in the fourth quarter.

Is there something in the way DVOA is built that is missing an indicator of Green Bay's dominance? I don't think so. Instead of looking at Green Bay with DVOA, let's look at Green Bay with a fairly simple measure: points scored and allowed. Most readers know that we can estimate a team's wins and losses based on points scored and allowed, known as the Pythagorean projection. And here's the thing: By Pythagorean wins, Green Bay is not the best team in the league this year... Green Bay's totals of 420 points scored and 262 points allowed work out to a Pythagorean win percentage of .754, or 9.0 wins. The difference between Green Bay's actual win percentage and Pythagorean win percentage is the second-highest since the merger, behind only the 1992 Indianapolis Colts.

I used the original article in 2011 to introduce our adjusted "Pythagenport" projections that adjusted the exponent in the Pythagorean projections to reflect the offensive environment that each team plays in. That raised the Packers' Pythagorean projection to .776. By the end of the season, the Packers had added on the loss to Kansas City and a big win over Oakland, so the gap between their winning percentage and Pythagorean projection had gotten smaller.

And where do the Chiefs stand now? This Kansas City Chiefs team has one of the largest gaps between the Pythagorean projection and actual winning percentage we've ever seen. Do you remember how we were all writing about last year's Packers and Seahawks? Everyone in the analytics world agreed: Green Bay was not really a "13-3 team" and the best evidence for that was their small point differential. That Packers team had a huge gap between projected wins and actual wins. And this year's Chiefs have an even bigger gap. Going back to 2000, the only teams with a bigger gap between Pythagorean projection and actual win percentage are the 2012 Indianapolis Colts and this year's Cleveland Browns.

Biggest Gap for Pythagorean Win Pct and Actual Win Pct, 2000-2020
2012 IND 11 5 0.688 357 387 -30 0.450 7.2 0.238
2020 CLE 10 4 0.714 368 374 -6 0.490 6.9 0.225
2020 KC 13 1 0.929 435 310 125 0.706 9.9 0.222
2004 PIT 15 1 0.938 372 251 121 0.719 11.5 0.219
2019 GB 13 3 0.813 376 313 63 0.611 9.8 0.202
2016 OAK 12 4 0.750 416 385 31 0.549 8.8 0.201
2009 IND 14 2 0.875 416 307 109 0.680 10.9 0.195
2011 KC 7 9 0.438 212 338 -126 0.254 4.1 0.183
2004 ATL 11 5 0.688 340 337 3 0.505 8.1 0.182
2019 SEA 11 5 0.688 405 398 7 0.511 8.2 0.176
2011 GB 15 1 0.938 560 359 201 0.764 12.2 0.174

I made this table back to 2000 so it would be teams everyone remembers, but we could take it all the way back to 1978 and the only teams that we would add above the Chiefs and Browns are the 1992 Indianapolis Colts and the 1982 Los Angeles Raiders in a strike-shortened season. Obviously, things can change in the final two games, especially if Kansas City sits its starters and takes a loss to the Chargers in Week 17. Still, the analytics community usually critiques teams that win this many close games. That happened with the Browns. That hasn't happened with the Chiefs.

The Chiefs are this year's leader in Pythagorean wins with 9.9, but the Ravens have 9.8 and the Dolphins have 9.6 and the Saints have 9.5 and the Packers have 9.2. Teams are pretty closely grouped together.

So, let's circle back to the issue I brought up a few paragraphs ago, that we need to separate the concepts of "greatness" and "dominance." I think it's fair to say that the Packers are a great team, even if they aren't a dominant one... The toughest tests the Packers will face will not be in their remaining four games. The toughest test will be the playoffs.

Despite what DVOA says, I agree with the conventional wisdom that Kansas City is the best team in the NFL going forward. Plenty of research suggests that offense matters more than defense and special teams going forward. It's something we're trying to figure out how to account for in future versions of DVOA. Count defense and special teams with 67% strength, using weighted DVOA to get more strength for recent games, and you end up with this top five:

KC 25.6%
GB 23.6%
NO 22.6%
BUF 20.5%
TB 19.6%

And with that change, we've made the Chiefs the best team in our ratings. This also doesn't account for the Saints defense getting to play against Kendall Hinton in Week 12. Account for that, and we've moved the Saints down below Buffalo. But then, we also should account for the absence of Drew Brees. He didn't have a good game this week, but his performance in Weeks 1-10 is more predictive than just one game that he had on Sunday. Take out the offense in the Taysom Hill games, and now the Saints are back to 24.8%, breathing down the neck of the Chiefs. Discount defense and special teams further, and the other teams move down but you do very little to change the gap between Kansas City and Green Bay. We still don't end up with a situation where Kansas City is ahead of the rest of the league by leaps and bounds.

The Chiefs are not exactly the 2011 Packers. There's no guarantee that they will lose in the playoffs in the same way. There are some significant differences between the two teams.

  • The Chiefs have better coaching than the 2011 Packers.
  • The Chiefs have a better defense than the 2011 Packers.
  • Patrick Mahomes' record of performing exceedingly well when the team is behind is like no other quarterback.
  • Based on weighted DVOA, the Packers declined over the course of the 2011 season. The Chiefs are not declining in a similar way.

We may not have the Chiefs as No. 1 in DVOA or even weighted DVOA but we still give them the best chance to win the Super Bowl because they're a very good team with a first-round bye. They win it all in 25% of our simulations, which is lower than other sites that do similar odds but still pretty impressive. The Chiefs even could rampage through the playoffs like an offensive version of the 1985 Bears, blowing out every opponent. It's a high-variance league, and that variance goes in both directions.

But I've been here before, nine years ago, with a very similar team. And that didn't end in the way people expected.

* * * * *

And now for something completely different... do you realize how bad the Houston Texans running game is this year? I keep meaning to say something about this. Right now Houston is averaging 3.54 yards per carry, not including scrambles. They have a league-low 36% success rate on running plays. It works out to -34.4% DVOA on runs, and that is the second-worst figure in our entire database. The only team that was worse? Last year's Miami Dolphins.

Worst Rush Offense DVOA, 1985-2020
2019 MIA -40.4%
2020 HOU -34.4%
1991 IND -34.0%
2016 JAX -34.0%
2005 ARI -33.3%
2013 BAL -32.8%
1986 NE -31.9%
2002 HOU -31.8%
2016 LAR -29.6%
1998 NO -29.1%
2015 IND -29.0%
2013 JAX -28.9%
2017 SEA -28.1%
2020 LAC -28.0%

However, it's not quite that easy. You might remember from a few weeks ago when I wrote about the New Orleans run defense that there's a bit of a quirk here because of the way DVOA is calculated. All plays, both passes and runs, get thrown into team DVOA. For offensive DVOA, this also includes some dead-ball penalties. The league average for all plays is set to 0% every year. Since passes are more efficient than runs, the league average DVOA for passes is going to be positive every year while the league average DVOA for runs is usually going to be negative. (Because all penalties are negative, it is occasionally slightly positive.) And the difference between passes and runs has gotten larger in recent years.

So instead of comparing the Houston running game to all plays, we'll compare Houston runs only to other runs. We'll do this by subtracting each team's run offense DVOA from the league average for that season. Now we get this list of the worst dozen running games instead. Houston is still on it, but much lower down:

Worst Rush Offense DVOA vs. NFL Average, 1985-2020
2019 MIA -40.4% -8.0% -32.4%
2002 HOU -31.8% 0.5% -32.4%
2005 ARI -33.3% -1.0% -32.3%
1991 IND -34.0% -4.8% -29.2%
1986 NE -31.9% -3.8% -28.1%
2013 BAL -32.8% -6.0% -26.8%
1998 NO -29.1% -2.8% -26.3%
2006 DET -26.5% -0.6% -25.9%
2000 ATL -24.6% 0.3% -24.9%
2016 JAX -34.0% -9.4% -24.5%
2004 SF -24.9% -0.8% -24.0%
2020 HOU -34.4% -10.4% -24.0%

That's still pretty bad though! Appropriately, the Texans have also run the ball much less than any other team. They have 241 carries this year, not including scrambles or aborted snaps. Jacksonville is at 272, and every other offense is over 290.

This isn't all David Johnson's fault, by the way. Comparing only runs to other runs, David Johnson has a rush DVOA of -5.8%, which is below average but not horrible. His running back success rate of 43%, using a slightly different definition for success rate, ranks 38 of 40 running backs. But Duke Johnson has a rush DVOA of -44.7% and C.J. Prosise has -120.8% on 10 carries and even Deshaun Watson is only at -0.2% when you take out scrambles and aborted snaps.

* * * * *

We'll finish up the week with an update on our odds of the NFC East producing a division champion with a losing record. Three of the four teams lost to non-division opponents, so these odds went way up. We're already guaranteed that the NFC East champion will be no better than 8-8. Going further than that:

  • Entire NFC East is 7-8-1 or worse: 71.9% (up from 43.2%)
  • Entire NFC East is 6-9-1 or worse: 15.4% (up from 4.4%)

* * * * *

Football Outsiders playoff odds, snap counts, and the FO+ database are now all updated through Week 15. A reminder that all our free stats pages, including DVOA and player position stats, now 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, and picks against the spread.

* * * * *

Here is the Football Outsiders Top 16 through 15 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 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.

WEIGHTED DVOA gives recent games more strength than older games to get a better idea of how well teams are playing now.

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>

1 NO 28.5% 1 30.8% 1 10-4 7.3% 9 -17.6% 2 3.6% 6
2 TB 25.5% 2 22.3% 4 9-5 14.0% 5 -12.6% 5 -1.1% 24
3 KC 23.4% 3 24.7% 2 13-1 26.4% 1 2.9% 17 0.0% 17
4 GB 20.0% 6 22.1% 5 11-3 26.1% 2 4.0% 18 -2.1% 26
5 PIT 19.0% 4 17.0% 8 11-3 -6.1% 22 -23.5% 1 1.6% 12
6 LAR 18.2% 5 20.3% 6 9-5 10.4% 7 -14.7% 4 -7.0% 30
7 BUF 17.9% 7 23.4% 3 11-3 13.7% 6 -0.2% 14 4.0% 5
8 IND 17.1% 10 17.3% 7 10-4 2.4% 12 -11.8% 6 2.9% 7
9 SEA 16.6% 8 15.4% 9 10-4 14.3% 4 4.2% 20 6.5% 3
10 BAL 16.4% 9 11.6% 12 9-5 1.6% 15 -7.0% 8 7.8% 2
11 MIA 11.9% 11 14.6% 10 9-5 1.3% 16 -5.2% 11 5.3% 4
12 TEN 9.0% 13 12.2% 11 10-4 23.5% 3 9.7% 28 -4.8% 29
13 ARI 6.8% 12 7.1% 13 8-6 1.0% 17 -6.2% 9 -0.5% 19
14 SF 0.6% 14 -4.4% 19 5-9 -4.4% 20 -5.6% 10 -0.6% 20
15 CHI -0.3% 15 1.5% 14 7-7 -11.5% 26 -8.7% 7 2.5% 10
16 WAS -3.1% 16 -1.0% 16 6-8 -20.1% 30 -16.2% 3 0.8% 14

If you're curious, the missing team in weighted DVOA (ranked 15th) is Cleveland.

For the full table, including variance, schedule strength, and non-adjusted VOA, visit the Football Outsiders DVOA database.


127 comments, Last at 04 Jan 2021, 11:13am

1 DVOA from Chiefs Saints game

The Saints weighted DVOA went up from that game? How? Did it really ever feel like the Chiefs weren't in full control?

34 New Orleans was winning in…

New Orleans was winning in the second half after the first drive touchdown and if they knew how to fall on the ball in the end zone it would've been 21-14 as opposed to 15-14. 3rd down variance was what changed the game. The Chiefs were like 1/18 on 3rd and long before this game where they seemed to get everything.

42 What if game

And if KCs backup punt returner calls for a fair catch it’s 14-14. Also if Matthieu catches that INT that hit him in the hands, KC wins 32-22. Lots of ifs, but KC was clearly the better team. They controlled that game.

63 Exactly

In reply to by nsheahon

And what if Breeland doesn't knock and interception out of mahomes' hand. What if someone actually tips his incomplete pass seemed like a fumble and the Wisniewski actually gets the first down? What if Mahomes sees the blitz and instead of getting sandwiched throws the ball away? 

2 My Problem with DVOA

can be stated simply. Of course one game sample sizes are, by definition, going to produce a lot of variance. But when we watch with our very own eyes Team A in control of a game but for a brief interlude to start the second half-- and then surrender a quasi-garbage TD near the end-- and DVOA concludes that not only was Team B better-- Ok, that conclusion is defensible if it's close-- but SIGNiFICANTLY better???  Something is wrong in the coding.  GIGO.

43 the sample size

The unit is a play, not a game.  A game has over 100 plays and that's more than adequate as sample sizes go. If  your eyes differ from DVOA, then your eyes and DVOA are measuring different things.

DVOA is very consistent about what it likes.  And it likes things that correlate well with football success.  

61 I don't know, I've seen a…

I don't know, I've seen a ton of games where Team A was "clearly" playing better than Team B, but failing to punch it in (or let Team B get back) and wound up losing.

At some point, you really want to be able to back up those subjective feelings with some objective data.

DVOA is good at sussing out some of those things that we tend to discount when just watching the game live. Things like those fumbles that Team A recovered and those fumbles that Team B didn't. Because a recovered fumble doesn't really *matter*, our brains tend to discount them. DVOA knows if a team has been lucky or nice.

That's not to say there aren't points to level at DVOA. My personal favorite annoyance is that it effectively punishes teams significantly behind for taking chances. E.g. a strip sack or interception thrown late in the 4th down multiple scores counts fully, whereas if the play works, it's discounted "because garbage time". That makes perfect sense if you are trying to correlate with win-probability, since throwing an int while down in the 4th is going to correlate very well with losing - and scoring isn't necessarily going to correlate with winning, if the team is down multiple scores late.

DVOA also cannot take into account that some teams will just play prevent if they are up big - and maybe even go a bit vanilla on offense - while other teams will continue to play 100%. DVOA (and most other advanced analytical play-based stats) compares the result of each play to the "expected" outcome of each play given down, distance, score, and time remaining. To some extent, this bias is alleviated by the D in DVOA, since adjusting for defense will help opponents of any team that consistently "plays 100%" late in games that are effectively decided.

123 Even the intro to DVOA says…

Even the intro to DVOA says those plays are "adjusted" and unless I'm severely mis-renembering, it's been described tons of times, so unless you can point to some specific official claim of the opposite, I doubt the veracity of your claim.

3 Broncos

What a weird year for the Broncos. They have somehow managed to go 5-9 against the toughest schedule even though they are the 3rd worst team. They also ended up with 5 players on the Scramble Digital All-Stars and 3 others mentioned as considerations and that's with their most likely pre-season All-Stars (Von Miller and Courtland Sutton) out for the year.

They must be cursed. They weren't that good to begin with and they've had to start a practice squad WR at QB, been killed by injuries and suspensions, had piss-poor play from the guy that was supposed to be their future at QB, and played a tough schedule. That should at least get you a top 5 draft pick but they may not even end up in the top 10. 

8 A year to forget in Denver

In reply to by wardh2o

Fangio has proven to be another Wade Phillips I'm afraid--awesome defensive coordinator; over-promoted head coach.  He's actually done wonders with the defense in general this year, since that's where by far the bulk of the injuries have hit.  But when the wheels come off on that side of the ball (Las Vegas, Buffalo) they are worse than the Jets.

Really wish I still had faith in Lock and Shurmur, and hate the idea of starting over again at either spot.  But #32 in DVOA on offense (yes, worse than Gase and Darnold!), when the injuries on that side of the ball really haven't been that bad (NFL normal, I think), is pretty damning.  I don't know how much is bad QB play vs bad coaching vs bad scheme, but it's all bad.

Of course, I expect them to now beat the Chargers and Raiders to finish 7-9, get a pick in the mid-teens and spend next year in purgatory as well.  And then what do you do?  Arguably at that point it would be a mistake to get rid of Lock or Shurmur and start over, and you would probably be better off praying that actual continuity at QB and OC for the first time since 2015, and some growth amongst the young players, makes them at least functional.  What a mess.

26 Bruce Arians?

Both former Colts assistants, FWIW.  Mora was fired/quit because he refused to fire Fangio as DC back in... 2002? 

Arians was a very successful OC in Luck's rookie year and led the team to something like an 8-1 record while Pagano was battling cancer.  (Pagano returned and the team lost.  Not his fault, but... cough, who's a HC these days. cough.)  That really put Arians on the short list for GMs.  (Did he win COTY?)  I would have preferred he replace Pagano, but Irsay wasn't returning my calls.

71 I can’t even blame the…

In reply to by Bobman

I can’t even blame the coaches for the wheels coming off the defence vs Buffalo, playing with pretty much no-one left at CB was always going to be really hard against a team with a good WR and a strong armed mobile QB. Lock turning back into a pumpkin after one good performance was fairly predictable. Maybe he will show signs of consistent improvement after a proper offseason, there are numerous excuses that can be made, but I fear they are more likely be shown to be just excuses rather than be shown to be reasons for slow development next season. 

If there isn’t someone they really want when the Bronco’s 1st round pick comes around I’d rather like them to trade back and stockpile for next year, the way college football has been this year probably makes this coming draft and even bigger crapshoot than normal, and if they can actually get healthy over the offseason and throw all the coaching and development help they can at Lock to see if he is capable of developing through hard offseason work (watching his opponent in Buffalo last weekend might inspire him as to what is possible if you really work on developing yourself), then if it fails they hopefully have the draft ammunition to go for a replacement in the 2022 draft.

Of course, I’m far from convinced that the Bronco’s management is capable of planning that long term. 

100 Of course, Arians' first…

Of course, Arians' first stint as a head coach was actually at Temple - in the 1980s, when he was in his early 30s! He went 24 years between head coaching jobs.

You have to trust the sources on that one, though, because the photos purporting to be of Bruce Arians at Temple don't look anything like Bruce Arians 2012-present.

95 Yeah I think most Colts fans…

In reply to by Bobman

Yeah I think most Colts fans wanted to keep Arians. They didn't want to deal with the PR hit of firing a dude they just hyped for an entire season for fighting cancer, but it seemed pretty clear. Maybe then they coulda also jettisoned Grigson and we could have ended that terrible phase earlier.


23 As a Broncos fan who hasn't…

In reply to by wardh2o

As a Broncos fan who hasn't been impressed with Drew Lock, I do think it would be a mistake to give up on him. Clearly something is wrong in the scheme, but I would love to see what they can do with the rookie receivers getting another year to develop and with Courtland Sutton back, and with an actual offseason of practice. Lock has accuracy issues and doesn't seem like a franchise QB, but I think the Broncos should stick with him next year unless a historic opportunity presents itself. Hey, maybe Carson Wentz just needs a change of scenery...

4 2011 Packers

it is an interesting comp. There were three factors, though, that may very well have contributed to the loss vs the Giants that DVOA would have no ability to measure.

1) Joe Philbin's son-- college age, I think-- fell through the ice and died in the Bye week-- I believe over that weekend. Easy to dismiss Philbin now for his failures in Miami, but he was the O.C. and very tight with Rodgers then, The entire team attended the funeral-- Philbin stayed with his job for the game that next weekend. You can't tell me it had no impact;

2) McCarthy rested Rodgers and most of the offense in the final week vs the Lions. (Yep- it became the Matt Flynn 6 TD game and made him some money form the Seahawks but, in a curious sense, might have helped Russ get to his eventual place as he had few expectations being drafted in the third round)  They weren't right in that Giants game after three weeks without playing, and while the Giants defense deserves some credit, it was (outside of the KC game Aaron notes) pretty much an outlier;

3) Rodgers #1 target was Greg Jennings back then. he was hurt sometime around Week 12/13 and didn;t play the last 4-5 weeks. So he and his QB had not had a game situation for at least a month and a half. there were a couple of key situations-- first half, i believe-- where the two of them misfired which could have kept a couple of drives going.

Doesn't mean the GB defense wasn;t suspect, or that Coughlin didn't outcoach McCarthy... But these factors were hardly inconsequential.

7 You can nitpick every team…

In reply to by oaktoon

You can nitpick every team that gets upset about what circumstances contributed to it. You can also do the same when the upset doesn't happen. "If only Eli hadn't twisted his pinky toe when chasing after his 2 year old daughter."

I will always hate these kind of after the fact rationals because they are purely speculative. Sure, they could be true, but its also more likely that shit happens in the playoffs and teams don't play well that day because they just didn't. There appears to be no causal explanation for why the Packers offense stunk up the joint against a bad Chiefs team earlier. 

27 Colts Steelers playoff game January 2006

If only Nick Harper's wife hadn't stabbed him in the leg the night before that playoff game, a back-pedaling QB wouldn't have run down and tackled a CB sprinting for a GW fumble recovery TD with nothing but green in front of him.

I play these games all the time, 15 years later, and they keep me up at night.  Any wonder I grind my teeth all night?  I probably need a shrink more than a dentist.  

5 2005 and 2011 both seemed…

2005 and 2011 both seemed like seasons where the heavy favorite was affected by outside forces (Packers OC and Dungy had both lost their sons in December, and their teams seemed understandably distracted).  Not sure I'd extrapolate from either example

6 People, including people who…

People, including people who read and post comments in these forums, completely struggle with the observed and the probabilistic.

The Chiefs could win the SB, might win the SB, are the favorite to win the SB, but that doesn't mean that A) Their DVOA is wrong and something in the formulas hates the chiefs or B) that they are immune to being upset even if they do happen to win it all.

I asked in one of these articles what people thought of the Packers in 2011 and the Chiefs this year because I too thought of these teams as very similar and the Packers, much like other all offense teams, is one bad day on that side of the ball from going home. For the Ravens of last year, it took a lot of turnovers and bad defense to send them home, not just one or the other. 

I often do this exercise. Before the playoffs begin, ask yourself which team you think is best and how does it stack historically. Because after the playoffs you are probably going to have a vastly different opinion and its more likely than not that that opinion will be wrong.

14 DVOA contributions

I always wonder how much individual game DVOA contributions are predictive. Without having the data, I feel that teams like BAL last year or NO and TB this year get a significant portion of their DVOA from blowing out inferior teams or a huge score from 1 or 2 big games. KC seems like a team that goes vanilla and doesn't want to show much when playing inferior teams. When they play good teams they go out and beat them rather easily. 

I don't think the formula is wrong or needs changed, but KC absolutely controlled NO during that game and a result that says NO was better just doesn't make much practical sense.

17 I dont think DVOA does this,…

I dont think DVOA does this, but dropped ints if they were factored, would definitely swing towards KC's favor. Brees could have thrown 4 picks in that game if it weren't for the drops. 

31 I agree. TB doesn't pass the…

I agree. TB doesn't pass the eye test. They blew away GB and it seems like that's where their high rating is coming from. I've felt for a long time like DVOA weights blowouts of bad teams too heavily, and every close win as a "coin flip."


18 +1 with a comment to comment 6 theslothook

Your point is right on target.  Pick a team to win the SB, any team.  Do it every year.  You will never get to 50% because taking the field is always the better pick.  If the Chiefs get the number 1 seed and only have to play 3 games to win the Super Bowl they would need to be 79% favorites to win each game just to be near/at the break even point.

.79 x .79 x.79=.49   This equates to being a 9 point favorite every game.  It would be difficult to have at least a 79% chance to win any one game against a playoff team, let alone all three.

For anyone that does not get the number 1 seed it is 84% favorites to win each game to be near/at the break even point.

.84 x .84 x.84 x.84=.50    This equates to being an 11.5 point favorite each game.  Thus, every team that has won a Super Bowl without getting the bye has defied the odds substantially.

Incredibly, from 1985-1999 the top DVOA team won the Super Bowl 10 of 15 years (will someone check my math, that seems impossible, I've gone through it twice).  The model was an incredible predictor.

Since 2000, only 4 of the top DVOA teams have won the championship.  We are in an age of parity.  It will be interesting going back in time with DVOA to the 1970's where some true super teams existed.

28 Patrick Mahomes laughs at mere math

Seriously, I did this with college admissions data for my oldest a couple years ago, 10 schools, acceptance rates ranging from 20% to 80%, the odds of X and Y and Z.  Ran it by an old college pal who is a consultant in this field.  She said "It doesn't work this way."  Knowing my son's grades/scores, she said "Any school that accepts 65%, he's a 100% lock.  Any school that accepts 25%, he's got maybe 1% chance.  So throw those ones out.  It's all in the middle."

That's what the eyeball test tells me about the Chiefs--if they are favored by 7 or more (or there is a corresponding DVOA spread), it's probably a lock.  Don't even include those in your matrix of possibilities.  1% chance of losing.  In your examples above, if KC had a 65% chance against all three teams (mathematically, well below a 50% chance to win it all), my guess is they're really probably closer to 65-75% likely to win it all.

Then again, Jets just beat Rams on the road.  

I always felt that individual matchups were more critical when two similarly stacked teams go head to head.  (Yes, Team A is better than Team B and facing the same ten opponents, A wins 9 while B wins only 8.  But Team B's all-world DT is going against an above-average rookie G who has been fine all season against above-average DTs, but not this week.  That up the middle pocket pressure just wrecked Team A's deep-passing game plan that worled all year.) 

And sometimes, that individual matchup is HC vs HC.

40 Chiefs magic spell on NFL fans - you speak the opinion of many

They won a championship by winning all of their playoff games after trailing by 10 points and have a 13-1 record this year.  This team is being talked about as if there is some magical aura, they will win all close games/they will win all games.  They will never get a bad bounce, a bad referee call, make a bad play with the game on the line.  Harrison Butker will make a 58 yard field goal in OT 99% of the time.

But ah, this is the playoffs they will simply crush everyone.  They were just kidding around early in playoff games last year when trailing by multiple scores such as 24-0.

They have been kidding around all season, barely beating the Raiders, Broncos, Chargers, Dolphins, Bucs, Panthers, and Saints.  But they were serious against the Jets, they just would not let the Jets hang around.

If you prefer a more concise version of my opinion, read comment number 15 below by theslothook.

I'll take the field, everyone else here can have the Chiefs.

And I will take the field against the 2011 Packers.

If I must pick one team to win the Super Bowl, I would take the Chiefs, simply because they are more likely than ANY other team to win it all, not that they are more likely than ALL teams combined to win them all.

Specifically I'll ask Bobman the Colts fan, if the Colts meet the Chiefs in the playoffs, what odds would you give the Colts to win the game?   Come on, your eyes would be glued to the TV set if this matchup occurs.

I will go first, I am a Ravens fan, I would give them a 35% chance of beating the Chiefs.  The Ravens have lost 3 consecutive games to the Mahomes Chiefs.  I will assume that the Ravens have learned to not blitz Mahomes on virtually every play, that would be a start to getting a victory.

The 1% chance of losing?  So they would go 99-1 when favored by 7 or more.  I'll take the underdog.  If you want to change your number to 10%, I'll take the underdog anyway.

The school comparison does not work since a school does not randomly pick the 65% or 25% of applicants that it accepts.

Using your HC vs HC argument I would match Frank Reich vs Andy Reid as an even match and Harbaugh vs Andy Reid as an even match.  These are 3 of the best coaches in the game today.  





45 KC Margin for Error

KC has a huge margin for error. That is why they have won 22 of 23, not because of “magic” or “luck”. They had a blown coverage for TD, a blocked punt for a TD against them, a muffed punt inside their 5, and two dropped 3rd down passes... all in the first quarter against HOU. And... they won by 20. They had a 4 turnover game on the road against a playoff team and won pretty easily.

They fumbled a punt for a safety, dropped two INTs, and lost the turnover battle on the road against the NFC 2 seed and still won comfortably. 

That is why people have confidence in them. They can lose any game. Any team can and the discussion of whether they are unbeatable is tired. No one is. But they continue to win in a wide variety of ways which makes them incredibly hard to beat.

110 Agreed

I was going to say they didn't lose the turnover battle but you're right they did. 1 Interception vs. Mahomes' fumble and Robinson's fumble.

The margin for error is huge. To a certain extent we are arguing semantics. The Chiefs are a great team no team can go 12 - 4 and win a playoff game and take the eventual SB champs to OT in the Conf. Finals, go 12 - 4 and 3 -0 in the playoffs and then 13 - 1 without being great.

But are they dominant? It doesn't appear so, if they win 3 blowout playoff games like the 86' Bears we will have to change our opinion but right now they do look beatable IF the other team plays a great game and the Chiefs have an average game.

It's hard to look at the 20 year career of Tom Brady and say there is no such thing as clutch play in the NFL. Mahomes has the exact same vibe with more physical talent and apparently the same drive to win, leadership and willingness to put in the work (physically and mentally.)

54 Colts vs Chiefs

Dude, I would LOVE this game.  I would also pee my pants a little, like a nervous yappy dog.

How would I feel about it compared to the 2003/2004 playoffs vs New England...?  I was pretty sure Indy could win them, and felt like they had an almost even chance in 2003.  Say 53/47% for the Pats.  In 2004, I felt Indy was better and had learned from their horrible experience the year before.  Maybe 51%/49% for Indy (reminder for the young uns, they lost both, and 2004 was worse for Indy than 2003).  In 2005 the top-seeded Colts lost in a mild stunner to the eventual champion Steelers that they had stomped a few weeks before--why was that not exactly shocking?  Because after the Colts first play in that game (an 80 yard TD pass from Manning to Harrison), it was basically a one-score game.  Yes the Colts led wire to wire, but the 2nd half was basically a draw and when you are an emotional psycho like I am, you are very sensitive to your adrenaline from play to play and during the second half of the reg season game, mine was not pumping.  I was depressed.  What the hell is wrong with them? Why aren't they dominating more?  The answer was, they weren't that much better, despite the score. So in 2005 when they lost to PIT by 3, a game they could have won if any of about three different plays had been slightly altered (a false start on a TD run at the 5 yard line after a 90 yard drive pushed them back to the 10 and they settled for 3 instead of 7, a CB's wife stabbed him inthe leg the night before and he happened to pick up the key fumble with a minute remaining and was tackled by a back-pedalling QB...)  That game was probably a 60/40 game that Indy lost because just about everything went PIT's way (except a Polamalu INT that was not called).

In 2006, I was playing with house money in my mind as neither team was expected to be there, but the Colts who had been repeatedly gashed by good ground games all season (300+ yards to the Jags in week 11?) had just defeated two very good running teams, in the Chiefs and Ravens.  The Pats also pulled a win out of thin air against the Chargers (who I felt could have certainly beaten Indy due to matchup issues).  So in that one, I'd guess I might have given Indy a 51% chance of winning. 

Against the Chiefs this year, I admit, I'd probably give KC about 55%/45% odds.  How does that break down as a spread, 3-4 pts?  Hell, a random bounce of a weird-shaped ball resulting in a turnover could equal 3 points in an instant.  I'd expect the Chiefs to win by less than a TD, but would not be shocked if the Colts won. Probably like most of the football watching world.

80 45 percent for Colts victory over the Chiefs?

In reply to by Bobman

“You had me from hello” makes for a much better movie line than “You had me from I would also pee my pants a little” but despite your hilarious but awful opening, I read on anyway.   So the math works out that the Chiefs win the championship 65 -75 percent of the time, unless they play the Colts, where they have a 45 percent chance of losing that game alone.

64 Ravens 35% Chance?

First off let me say I was a big Eagles fan during the Reid era (now I like KC better - I followed Reid over) - so I obviously I respect and root for Harbaugh and I like Jackson's "break the mold" talent.

I don't disagree that the Ravens certainly would have a chance against the Chiefs and 1 in 3 might be a good guess. 

However, If I were a Ravens fan I would be a little disconcerted that it is the Chiefs that have learned the Ravens not vice versa.

The first game Ravens had the lead and with less than 2 minutes left and 4th and 9 happened. They still almost could have pulled out when Mahomes fumbled in OT but was alertly covered by an OL. Even then had Lamar not been injured (?) and been in for the last drive instead of RGIII they still might have won or salvaged a tie.

Game 2 the Chiefs blitzed the Ravens and were out to a big lead at half (24 - 7 or something like that). The Ravens came back eventually cutting the lead to 5 but they never got the ball back as the Chiefs killed the last 4 minutes of the game on offense.

Game 3 the Chiefs won by 14. The Ravens never had the ball with a chance to even tie in the 2nd half. And of course were aided by a KO return TD.

The games are going more in the Chiefs favor not less. I'm pegging Baltimore's chances more at 1 in 5. I love Lamar but he has been average at best in big games against quality opponents and you need more than that against the Chiefs.

44 hmm

Do it every year.  You will never get to 50% because taking the field is always the better pick.

There's no mathematical or physical reason to think this.  Be careful with words like "always".  

Whether the field is the better pick (and it usually should be) depends on the gap between the favorite and the rest of the pack.  Handicapping the Super Bowl just isn't as hard as you make it out to be.  There is no a priori reason to think that "the favorite" cannot win more than half of the Super Bowls over a given stretch of years. 

46 It's passing. The rules have…

It's passing.

The rules have so tilted towards pass offense that it's nearly pointless, and nearly impossible, to build a true super team anymore. So the greatest teams are prone to being cut out by one bad game, or the opponent's QB going on a high-variance hot streak.

In an era where overall team build mattered, you saw less of that, because one guy going off or falling off mattered less.

84 Intended effect?

I'm pretty certain that the intended effect of the salary cap is to make sure the customers are on ownerships side during contract disputes. It had that effect immediately, even though salary cap hell is a fictional place. 

13 The DVOA Model Is Clearly Broken rn

I'm a long time KC fan, so I'm certainly biased, but there are at least three reasons I just don't think DVOA is reliable when it comes to evaluating the 2020 Chiefs.

1) No preseason - The FO model has a very precise way of discounting early-season games based on how predictive those games USUALLY are. Those estimates are based on several decades' worth of years with preseason games. We didn't have a preseason this year. Does that make early-season season games less predictive, or more? No one knows. There's a major confounding variable this year, which cannot be adequately controlled for, and which makes all historical data-based modeling less reliable. 

2) 3rd downs - Hopefully we can all agree Patrick Mahomes distinguishes himself most from other passers on 3rd down. He gets conversions he shouldn't get. I've heard it said - maybe I have this wrong - that 3rd down conversions are not usually considered predictive of success in the playoffs. If Mahomes doesn't follow the typical laws of gravity for 3rd down conversions for some reason, and it certainly seems he doesn't, it makes sense that historical models would underrate his brilliance, as pretty much everyone but Aaron agrees that they do. 

3) Offensive red zone - The Chiefs 2020 offensive red zone is one of the craziest things I've ever seen in my years as a football fan. They were actually pretty meh last year in the zone. This year, they were amazing. Then they won the division and after that they were meh again. Then against the Saints, back to amazing. It's not just success vs. failure either. Anyone watching the Saints game could see Reid and Bienemy have a deep well of trick plays they pull out only when they need to. You didn't see that stuff against the Broncos this year. They saved it for when it really mattered. As with 3rd downs, even if historical models were able to predict future playoff red zone success - I never heard anyone say they can - there is no way you can mathematically account for Reid and Bienemy's trick play game. For that, you have to use your eyeballs.

All of which is to say that there are some solid, objective, mathematical reasons to ignore the models and use some good old common sense. Even my 8-year-old nephew (a Broncos fan) knows that the Chiefs are currently dominating the NFL.

15 You really have to believe…

You really have to believe in this theory that Chiefs are intentionally playing possum to think they are dominating the NFL. They are certainly the best team in the NFL, but this isn't some 1980s 49ers level domination we are talking about. 

24 yes, that's what I'm seeing

Case in Point: During their red zone slump, they had a play where Kelce tried to throw a TD pass to Mahomes. After it inevitably failed, it came out that Mahomes designed that play himself.

If you've ever played a game of Madden, you should already know that once the other team starts trying to throw TDs from TE to QB, they've taken their foot off the gas.

30 Or they are incredibly desperate

I can recall a Colts game vs the 49ers in 2009, a SB year for them, and just weren't getting it done.  They trailed for three quarters and had no TDs up to this point.  Would you EVER take the ball out of Peyton Manning's hands for a pass play?  Probably not.  Like 99.9% not, right?  And yet...He handed to lefthanded Joe Addai (a HS QB, everyone found out after the fact) who ran left, then stopped and passed for a 15 yard TD to Reggie Wayne.  Their only TD of the day and a very closely contested game was over. (How does Manning pass for 350 yards and no TDs?)

Not sure they ever used it more than once or twice again, but they apparently felt desperate enough that trickery (in an otherwise very vanilla offense) was what they needed.  Since Manning had a reputation of getting three plays called in and he could select (I've never heard him, Tom Moore, or Dungy say that, but everyone accepts it as fact), or audible if need be, I assume that was his call.  And the right one.

16 2011 Packers

No QB ever threw the ball better than Rodgers in 2011. He hit 6-inch windows, back shoulder throws all season long. Unbelievable.

In Week 16, McCarthy sat Rodgers the second half. In Week 17, Matt Flynn made millions throwing for 6 TDs and 400 yards against a bad Detroit defense while Rodgers again rode the pine. Add in the bye week and the Packer offense and Rodgers hadn't played together for almost 3 weeks. On a cold, crappy winter day in Green Bay, the Giants pulled the upset as the Packer offense couldn't get in gear.

McCarthy was a stout adherent to the 56-minute game philosophy. Even if the offense had racked up 400 yards in the first 56 minutes, the final 4 minutes were devoted to burning the opposing team's timeouts rather than converting first downs. Ideally, this was done strictly by running the ball. McCarthy routinely turned 10 point 4th quarter leads into nail biters.  This 56-minute approach cost the Packers a trip to the Super Bowl when the Seahawks rallied in the last 5 minutes. 

If the Packers offense is in sync, they could give the Chiefs a game in Tampa. The Chiefs would be favored. The Packer offense would have to be mistake free. It'd be a great matchup!

55 "No QB ever threw the ball…

In reply to by PackerPete

"No QB ever threw the ball better than Rodgers in 2011. He hit 6-inch windows, back shoulder throws all season long. Unbelievable."


 The problem is, everyone has a bias. I freely admit to mine. I have met older Packer fans who will swear Favre at his peak was better with some (very) subjective era adjustments. For what its worth, 04 Manning(with heavy bias) and 2011 Rodgers were the finest displays of QB play I have ever seen.


As someone who is now a football historian, I find the Rodgers Favre comparison delectable. Sure, by raw comparison, Rodgers is better but how much to era adjustments work in his favor? I have asked packer fans this question and surprise, teh answer correlates with age - with Displaced as the notable exception 

58 Rodgers is more reliable …

Rodgers is more reliable (Favre was bit like a hand grenade -- something exciting was going to happen, but it wasn't always good). but it feels like his precision has shaved the edges off his genius a bit. Favre was high-test -- it made things happened, but burned going down.

78 Hot 1990s take

I think you have to see it in context too. Favre was almost unarguably the top QB in the league from 1995-97 – his best years – whereas Rodgers has always played simultaneously with Brady and Brees (and, for some of his career, Manning). Favre was doing things that no other QB was doing at the time. I mean, I agree with your opinion, but the mid-1990s were a downtime for quarterbacking, and I think you have to evaluate play in those circumstances. In which case, the distance separating them is reduced, at the very least, I'd say.

82 The argument in Favres favor…

In reply to by ammek

The argument in Favres favor comes down to era and receivers. It's indisputable that passing is much easier today than it was in Favres time, so it becomes a question of to what degree? 


As for receivers, I didn't see the meat of Favres zenith years so I can't comment one way or the other.

19 Weighted DVOA vs DVOA and championships

I went back and have found that 10 of 15 teams won the Super Bowl that were number 1 in DVOA from 1985-1999.

Since 2000, only 4 such teams.

But number 1 weighted DVOA teams only had 7 championships from 1985-1999, and only 3 champions since 2000.

That makes 14 DVOA champions vs only 10 weighted DVOA champions.  This may be a random variance, but there is some sense to the idea that it is better to play the best over an entire season since that team is more likely to get a bye.


20 As a huge Packers fan, and a…

As a huge Packers fan, and a now 7 year resident of Kansas City the 2011 Packers and 2020 Chiefs comparisons are very solid. Though the Chiefs are not nearly as flawed. I'll be using a lot of conventional stats because that's what I have access too, but the 2011 Packers were a great passing offense with other parts thrown around it. KC 

Though I find it hard to believe the 2011 Packers had a better running game than the 2020 Chiefs. That 2011 squad had Ryan Grant, and James Starks and never had a 100 yard rusher. They had 8 games where they hit 100 yards or more but the "dual headed" running back was not good. KC already has 8 games (with 2 to go) of over 100 rushing yards, and one game over 200. Even with teams needing to defend James Jones, Greg Jennings, Jermichael Finley, Randall Cobb, Donald Driver, and Jordy Nelson you would have so many times where Grant or Starks would have 7-10 carries and 10 - 20 yards total on those carries.

The 2011 Packers were a great QB with 5 good to great young WR and end of career Donald Driver. I just can't buy they were a significantly more efficient running team than the 2020 Chiefs. I know running efficiency seems to continually go down, but 2011 GB being positive? Grant and Starks were 46 and 49% success rate players, ranking in the 19th and 30th spots. Edwards-Helaire is at 56% and 11th. Maybe the last 2 games of the season will pull those ratings closer. I'm not saying KC is a good running team, but my memories were that Grant and Starks could only do something decent when there were huge holes because every pass catcher on that team needed to be covered. I recall lots and lots of stuffs and runs for losses, and games where one of them would have one run that accounted for like 50% of their total and then 7-8 runs that got maybe 2 yards so they would have a 3.8 - 4.4 ypa, but it was still a horrible game.  

The 2011 Packers defense was horrendous. 2020 KC has given up 300 passing yards twice this year. 2011 GB, 7 times, 400 once, and 500 once. 2011 GB which was an amazing forerunner team, so opponents were behind and passing more, also managed to give up over 100 rushing yards 10 times. 2020 KC is only at 8. 2011 GB gave up 400+ total yards 11 times. KC has done that 5 times. Both teams played against -2 to -2.2% DVOA schedules. I'm amazed the differences aren't even higher in the DVOA for the teams.

I loved that 2011 team but it was the most one dimension team I've watched. McCarthy "forced" a run/pass balance as I recall, the passing game was so fun, but the offensive flow could get so mucked up by a run into the LoS. It didn't matter most times of course, until it did. Even still the 20 points they scored on the Giants, despite 4 turnovers was the most the Giants gave up in the playoffs on the way to the SB win. 

Anything can happen, but 2020 KC is in much better shape than 2011 GB. 2020 KC has a slightly worse passing offense because while the top 2 receivers (Hill and Kelce) are better than where Nelson and Finley were at that depth of that 2011 GB was nuts. 2011 Jennings was better than 2020 Watkins, Hardman, or Robinson. And 2011 Jones and Driver are on the same level as those 3. Then you still have rookie Randall Cobb. DVOA reflects that too. It's close but I see why DVOA gives the nod to 2011 GB passing. Add on that despite what DVOA says, the 2020 KC running game is better than 2011 GB running. The 2020 KC defense is clearly better, and 2020 Andy Reid is clearly better than 2011 Mike McCarthy at coach too. So yes 2020 KC while similar to 2011 GB is a better team and less likely to be tripped up.

21 And yet, for all of what you…

And yet, for all of what you said, I think the Packers in 2011 looked like a tougher out than these Chiefs.  I posed this exact question, if these two teams met on neutral field, who would be favored? And to me, that question really came down to how much does a passing offense trump everything else and how much do you believe the Chiefs are saving their best stuff only for when they are threatened.

I will turn it around say Vegas believes pass offense is the trump almost all the time. The Broncos in 2013 were favored despite being a severely deficient team everywhere but the pass offense. The same was true for last year's Chiefs who really had a dominant pass offense and a goodish pass defense but were terrible against the run vs the overall strong 49ers. It didn't matter.

I think, me personally, I'd favor the Packers because their pass offense really would be the difference. 

35 I might take the 2011…

I might take the 2011 Packers of the 2020 Chiefs too. They are mirrors and the 2011 Packers gave up more in other areas to make the best dimension stronger. In that situation I'll generally go with the team that has the strongest single dimension.  

But the question is more would you take the 2011 Packers over the field or the 2020 Chiefs over the field. As has already been pointed out, the field is pretty much always the better bet, but your choices are 2011 Packers or 2020 Chiefs. Ignore hindsight that shows the field beat the 2011 Packers. In that case the Chiefs being better at everything else feels like it minimizes their chances of slipping up. Their passing offense is already past the threshold of being able to beat anyone in the field. But their defense also has a chance to stop someone, their run game also has a chance to make a play when needed. 

Heck the 2020 Packers or the 2011 Packers vs the field is an interesting question. The 2020 Packers haven't quite hit the threshold where their passing offense can beat anyone. I'm not sure what that is, but they aren't there because they don't have enough weapons. Carolina just showed eveyone how you stop the pass you take away Davante Adams and get enough pressure on Rodgers to slow him down. The 2nd and 3rd options aren't good enough to beat you on their own. But the 2020 Packers can lean on the run game too which they did to beat Carolina. The 2 games they played without Adams earlier in the year, NO, and ATL they won because Lazard wasn't injured yet against NO and was able to take up Adams role and against ATL there wasn't enough pressure to keep Rodgers from carving them up with the RB and TE positions.

The 2011 team couldn't do what the 2020 team did. When Jennings was out against KC and Tamba Hali was getting pressure they tried to use the RB, TE, and Cobb in the slot to carve them up. But KC kept good coverage on Finley and Nelson and they held onto the ball  (36 to 24 on time of possession). Grant was running decently but only got 12 attempts. KC kept Nelson covered so Rodgers didn't throw to him and they made all the throws to Finley (who caught 3 of 10 targets) tough. They basically played the game Carolina just played against the 2020 Packers. The 2020 Packers have a better coach and more options though. While they did not play well, the defense was good enough to force Carolina to punt 4 times. The 2011 defense held KC to FG's but they let drives keep going forcing 2 punts. They couldn't stop anyone. Just like the 2011 defense couldn't stop the 2011 Giants who only punted twice. And the Giants played coverage to take away Jennings and slow Nelson. They got pressure.

The 2020 Chiefs can do more than pass. Their defense can force the other team to punt from time to time. They are very similar to the 2020 Packers. So close I remember a few weeks ago that I mentioned that the Packers were ranked one spot behind the Chiefs in nearly everything. That's diverged some as GB has gotten progressively worse on special teams and KC is getting better on pass defense. KC's 30th ranked rushing DVOA is still a 2.6% which means that teams are barely better than replacement level rushing against them. Their 13th ranked pass defense is 3.1%, again similar. But that matters way more when passing offense is the biggest thing to worry about. That 2011 GB defense was 4.0% against the run and 13.3% against the pass. The 2020 Packers are -4.6% (so actually OK though ranked 21st) against the run and 10.2% against the pass so noticeably better than the 2011 Packers. But also noticeably worse than the 2020 Chiefs.

The Vikings and the Colts exploited that pass defense. Minnesota didn't do a lot of volume because the Packers couldn't stop Cook but where stupidly efficient when they threw and they did need to throw or the Packers D would have stopped them a few more times. The Colts simply stayed very balanced picking apart both weaknesses and taking advantages of turnovers. I think the 2020 Colts could have beat the 2011 Packers in the same fashion though it would have been a higher scoring game for both teams. I don't think the Colts would have missed the FG because they would have scored a TD instead and I don't think they would have punted at all. But the 2011 Packers would have scored more too. The 2020 Vikings would have had the same type of shootout games they had this year with the 2011 Packers. Though you wouldn't have seen as many of the long drives the first half of the 2nd game had. 

So against what the Field has to throw at the teams, I think the 2020 Chiefs are less vulnerable than the 2011 Packers. I think the 2020 Packers are actually less vulnerable than the 2011 Packers were. I actually think DVOA is doing a good job. The 2011 Packers were a 26.2% DVOA team, had a 23.4% weighted DVOA. 2020 Packers are 20.0 and 22.1%. The 2011 Packers played the 29th rated schedule the, the 2020 team should end up around 19th (currently 23rd with the 13th toughest future schedule). I think the 2011 Packers would have lost a game sooner, and possibly a 2nd game if they had faced a tougher schedule. That's actually my biggest worry about 2020 KC. They've played the 27th ranked schedule and the last 2 games are ranked 29th, another 2011 GB parallel. Calling all three of these teams 13-3 quality teams with 2 of them being a little luckier than the third doesn't feel wrong to me. 

Though I do need to get a better sense of DVOA and win loss correlations. 20% feels like you should be an 11-12 win team. 25 is 12-13, and that 30+ is where you should be seeing the 14+ win teams. But I haven't actually ever looked at the data.


Edit: Another thing about that 2011 Packers team...

Which team has the most points per drive? 2020 Chiefs, 2011 Packers, or 2020 Packers? 

Answer it's a tie both the 2011 and 2020 Packers score 3.05 points per drive. 2020 KC is 3rd this year at 2.91.
The 07 Pats were 3.37.
The other over 3's since 07 are
2016 ATL - 3.04
2018 NO - 3.21
2018 KC - 3.25
2019 BAL - 3.08

Now that is more an argument for the 2011 Packers being superlative since adjusted the 3.05 in 11 is better than 3.05 in 2020. (and that juggernaut 07 Pats offense....). But as has been stated the 2018 Chiefs were a better passing offense (63.5%) than the 2020 Chiefs (53.4%) and just behind the 2011 Packers (67.8%).  The 2018 KC team lost 5 games (4 regular season, 1 playoff) with a better overall DVOA, but defense much closer to the 2011 Packers than the 2020 Chiefs. It's just another point that even being really amazing at the most efficient thing in football doesn't mean you are unstoppable. The 2018 version of KC went 12-4, underperforming their DVOA (33.2%) a bit. The 2020 Chiefs are 13-1 over performing their DVOA (23.4%) a bit. The 2011 Packers went 15-1 overperforming their DVOA (26.2%).

Again I loved that 2011 team and coming off the SB win it was such a fun season. This 2020 team though has been one of my favorites to watch. The 95, 96, and 97 teams still hold that crown they were even more fun because they were better balanced (and Reggie White should have gotten the Super Bowl MVP not Desmond Howard). But 2020 has started to play enough defense lately that it doesn't drag the experience down.   

32 The 2010 Packers defense was…

The 2010 Packers defense was playing great, especially at the end of the season. However, starting FS Nick Collins was injured in September 2011 and the defense was a sieve against the deep ball after that injury (despite Charles Woodson). McCarthy's habit of turtling at the end of games wasn't the only problem; so was the defense giving up a big play on every other deep ball when the opponent was trying to come back. Usually they'd intercept one to save the game, but not before allowing the game to get close in the first place. 

53 Injuries rob of us so much…

Injuries rob of us so much potential greatness. Nick Collins was on a multi-year All Pro trajectory. That dude could play. Woodson was 35 years old and I think Tramon Williams was still in his "Admiral Armbar" phase too. They did what they could but Collins was the glue in the Capers scheme. His scheme needed a minimum level of secondary play or it would fall apart. A great safety could make that minimum skill requirement lower. Morgan Burnett met that minimum for a few years as did an aging Woodson and Tramon Williams after a few seasons was fine as well, so was a healthy Sam Shields. But Charlie Peprah, MD Jennings, and a rookie Tramon Williams, and Sam Shields injury replacements (because Sam was injured too much) simply didn't. It was part of the huge variance that D saw because Capers wouldn't or couldn't adjust his schemes to the talent he actually had. When the minimum requirements were met it was a good scheme. When they weren't the gaps that could be exploited were huge and safety was one of the few positions that could help cover for deficiencies. That applies universally of course, but it was especially true of what was asked of a Caper's secondary.

The other what could have been career destroyed by injury that jumps to a Packers fan's mind is of Sterling Sharpe and the HoF trajectory he was on. He is still the best Packers receiver in my head (which is basically just the guys who played during my lifetime because personal experience trumps stats when thinking about players). Lofton played during my lifetime, or course, but I was so young when he was a Packer that I really only remember his last season and then his time with the Raiders and Bills. I will not argue with someone who says Lofton was better, but for the players I can remember watching it's Sterling Sharpe. For my lifetime (which started in 75 I'm not that old but I'm not young anymore either) I put them Sharpe, Lofton (mostly on stats, reputation, and what I saw of him on other teams), Adams, Nelson, Jennings, Driver (longevity), Freeman. Then it gets all muddled with the Jones/Cobbs/Brooks/Walker/Schroeder tier. If Lazard comes back from this core injury and plays like he did at the beginning of the season for a season or two he could get in there on his own; I really do enjoy watching him when he's fully healthy.

Darn injuries. Every team has players like that I'm sure. But those two. LeRoy Butler was cut short too, but he still had an 11 year career so I don't lament that as much.  

Before anyone gets into it I of course will never argue with someone who says Don Hutson was the greatest Packers receiver. Anyone who puts 74 rec, 1211 yards, and 17 TD in a 14 game season back in the 40's is clearly a stud WWII caveats or not. Anyone who leads the league in rec 8 times, yards 7 times, and TD's 9 times, and yards per receptions twice (in years he lead the league in receptions no less) in an 11 year career is also a stud. I've seen stories of him running a 9.7 100 yard dash. I that would be around 10.2 - 10.4 100 meter assuming he could keep his top speed for another 9 meters which would be fast enough to win all the Olympic 100m races through 1960, as well as the 1980 Olympics (which the US boycotted so they were a bit slower). I'd like to think someone with that speed and obviously with decent hands magically transported to the modern era and given a few years of coaching could still play the modern game. But of course I've seen only some of the old film of him.  

73 Regarding Lofton

From 1980-1985 averaging 19 plus yards per catch with multiple defenders hanging all over him as the Packers only legit talent on offense being thrown to by the likes of David Whitehurst and Lynne Dickey is one h8lluva an achievement. 


Oh, and head coaches Bart Starr and Forrest Gregg.  


The football gods did not do James Lofton's career any favors is all I am posting.  

74 And the single greatest catch I ever saw

In reply to by big10freak

was Lofton converting a 4th and I think 21 by catching the ball one-handed as he was being dragged to the ground by three defenders including one by the facemask (no penalty of any kind called on the play)  Think GB still lost the game.  But not for lack of Lofton's effort.  


I need to sift through the play by play history and find that one

77 Knocks against Lofton

I barely remember Lofton in green and gold, but the Packer fans who suffered through that era sometimes hold it against him that 1) Lofton took plays off; 2) his hands were inconsistent, in that he'd make one spectacular catch and then drop a far easier one; and 3) he was a bit of an ass.

Sharpe was the most complete Packer receiver I remember, although the Ted Thompson picks brought a glut of very, very good WRs.

101 I know you have to take some…

I know you have to take some of those old stories with a grain of salt, but it would explain a lot if Don Hutson was truly one of the fastest sprinters of his generation and not just extremely fast for a football player. No mention of him ever being a track athlete on his wikipedia page, though.

39 Rushing

Chiefs have 1 more fumble on 30 less carries. That and the Chiefs RBs having higher success rates but lower DVOA means they're running (usually) shorter yardage situations where they get exactly as much as they need iirc

Like I've stated a couple times I'm taking the '11 Packers. Chiefs are the overwhelming favorites now because, yes they're good, but the rest of the league is...meh. The Packers had legit competition in 2011 in the Patriots and Saints. But then the one game elimination postseason started and anything goes. Those three 2011 team (and the 49ers believe it or not)  all had a higher SRS than the current (first in the NFL) Chiefs. Which matches the eye test of their games being kinda close and, yep, they're 7-7 against the line. 

Then other small things like the Packers only regular season loss being on the road by 5 as opposed to the Chiefs losing at home by 8 etc

111 Rushing Fumbles

In reply to by ImNewAroundThe…

The Chiefs have exactly one fumble rushing the ball. Darwin Thompson. Their QBs have a few fumbles (but I don't believe any came trying to rush) and so do their WRs but Running Backs have been virtually fumble-free. After his fumble in Game 3 (I believe) Thompson has barely seen the field on offense.

25 Drive based DVOA

The unintuitive result of NO besting KC in DVOA this week could, in my opinion, be fixed by applying DVOA on a per-drive level instead of per-play. The Chiefs sustained long, successful drives by running a ton of plays. This dragged down their yards/play and made them (incorrectly) look inefficient. The Saints did the opposite; they had a bunch of failed drives that ended quickly and a handful of big plays, which artificially boosts their yards/play.

Aside from a turnover, going 3-and-out is the worst possible drive outcome. But DVOA barely penalizes such a failure because there aren't enough plays to make a dent in the overall average. NO didn't get a first down until midway through the second quarter, but their play count was so low that a couple good drives later more than canceled out these major failures.

We also saw this in week 1 with MIN having a better DVOA than GB despite getting dominated. We saw this in the 2015 playoffs when CAR led SEA 31-0 at half but had the lower DVOA because SEA didn't run enough bad plays when the game was actually in doubt. It's an issue that's long overdue for a fix.

48 Excellent point

This is an excellent point. The per yard looked relatively decent for NO when they went 3 & out numerous times the had a 50 yard pass play. But they weren’t threatening much at all and they really moved it against KCs soft coverage up 2 scores with under 4.

56 Per Drive Stats

I love the per drive stats here and was just perusing them for the Colts.  I was surprised by a few things and discovered a potentially tragic flaw that the right matchup, even with an inferior team, could spell their doom.

Their offense is surprisingly higher ranked than their D in terms of almost all drive stats.  (Earlier in the season, say six weeks ago, their D was a top unit and their O was ranked about 20 overall by DVOA--that gap has closed somewhat.)  Their ST (&D) really set them up well in terms of starting LOS (ranked 1 and 2), with most rankings between about 5 and 15.  They're a solid team, except for D time of possession (I think that's the one) which--in simpler terms--means 3rd down efficiency.  Maybe 23rd ranked or so.  Since they have a very good run D (which usually corresponds to low TOP), I assume their TOP flaw is third downs and allowing the opposing QBs a high completion percentage.  If they face a team that just has a knack for converting, or more likely usually ends up in 3rd and short due to O efficiency and good situational play calling, the Indy D will never get off the field unless there's a TO or a score.  Their TO luck and skill has been pretty good this year, but I hate to rely on it (like the last two Texans games.  gulp).  So I can reasonably see them being beaten consistently by a team ten DVOA slots lower, if that team matches up where they are weak.  

Every team has a weakness and it might show up on the drive stats page.  If you can find a reasonably competent team that has their strength where your one team has a key weakness....

I didn't go back over the two Raiders games for KC, but I would not be surprised if that is part of it--they're a good matchup for LV for some reason.

Gosh, but I love this game. I just wish it didn't take such a toll on the players. And I'm 55% glad my oldest opted to not play in college (though I wish he'd opt to get his torn labrum fixed).

33 RE: some comments here about the Packers rest in w17 2011

And that the "rust" was the (or at least a big) reason they lost. In reality I dont think that's real or real enough to significantly sway anything. Maybe I read them wrong but it felt implied. 

Imagine he plays that game and he gets hurt, not only is that season in jeopardy but maybe his whole career. *shudders* In fact that happened one week 17, 7 years later, except it was the opposite intentions, they were already eliminated and Rodgers got "only" got concussed AGAIN. And they had just reinvested in him again with a record breaking contract. Imagine he got Joe Theismann'd. "Culture" (of winning, when the season is dead in the water) wouldnt explain the hysteria that'd follow.

In hindsight it may seem wrong but it was the right process. And with the right process that Packers team beats the Giants 9/10. Unfortunately (for me) that 1 time was reality. 

And like I stated the other week I'll (still) take em over the 2020 Chiefs even though they are similar. Dont count your SBs before they hatch Chiefs fans. Even with a supremely talented QB. I remember sitting on top of the world blowing that Giants L off like "we got it next year". Havent been to the SB since. Although this year...still nothing is guaranteed. 

36 I feel like the "Chiefs are…

I feel like the "Chiefs are the Warriors of the NFL" (a trope that is magnified by Mahomes' similar  appearance to fellow second-generation pro athlete Steph Curry) expression is fitting here.

I remember the 73-9 Dubs being "so good with their high-flying offense they didn't have to give the starters that many minutes and could take the foot off gas" and we saw how that turned out.

I expect KC to tear through the playoffs now with Pit slumping, but those unexpected slip-ups will happen plenty of times in future years, especially when Mahomes' payday leads to cuts elsewhere. 

41 I expect the same

But then again if you told me the Steelers (or someone) left KC with a 3 point victory in the postseason I wouldnt be too extremely shocked as that is in the realm of possibilities (even if it looks impossible right now). Thought the same when the silly 9-7 Giants came to town. Ugh. 

This season really is up for grabs but every year will still have it's favorites. 

50 I would. The Steelers are a…

In reply to by ImNewAroundThe…

I would.

The Steelers are a terrible tactical matchup against KC. It's a blitz-heavy defense that can't run the ball and doesn't throw deep any more.

Let's face it, the 2019 49ers were a team custom-built to take down KC.

103 Think it has more to do with Carr throwing 3 TDs

and getting 347 yards (7 more than Mahomes) on 22/31. A 11.19 Y/A, 11.68 AY/A, 126.7 passer rating is what was brutalizing. Rather than "just" grabbing a lead (Chiefs aren't known for making comebacks at all) and expecting an average of 4 ypc to carry you to victory over the next 3 quarters. TOP didn't help LAC, New England, Denver (twice), Carolina, or Miami.

If you're main objective is to run the clock out (outside the 4th quarter) just to "brutalize" TOP then you're likely gonna have a lot of 3 and outs/short drives. More nuanced than that. Because the Chiefs pass D DVOA is 3.1% compared to their 2.6% run D

108 It did help. Just not enough…

It did help. Just not enough.

In October, the Raiders were +11 in TOP despite being +1 in plays. They ran 15 more rushes, for 144 yards. They had 8 rushing 1st downs. They were 7-14 on 3rd and 2-2 on 4th. They drained the game on a 19-yard drive with 7 rushes and 1 2-yard pass.

They passed efficiently. Ran efficiently, setting up 3rd-shorts which they could reliably convert. They balanced turnovers. They turned red zone trips into TDs. KC still averaged 3 points per drive. LV still had to out-score them. It's still hard to beat a 13-1 team. The rematch game was similar, except KC was a little better at everything and LV a little worse.

Only one team had successfully launched a pass-heavy shootout with KC and won, and that took 5 turnovers, including 3 INTs. KC still scored 51. Mahomes has one other 3 INT game. They won that one.

There's no perfect model. But you can stress KC's offense by attacking their defense, and not giving them any room for error. They can still win that way, and often do. Even NE, who jumped out to a huge lead and ran all over KC and won TOP by like 37 minutes and 55 plays, still only got to OT by having a game-losing INT called back on an offside, and only won because of the vagaries of a coin-toss. That team won a SB over the other team to win a shootout with KC. There's no sure-fire model. But KC likes getting into pass-heavy shootouts. It's what they are built to do. Bad teams can't begin to hang with them that way. But they can hang by trying to drag KC into the mire. They often lose, but they were going to lose anyway. But some win this way.

116 The Raiders game was close through out.

Yeah, they ran a 19 yard drive with more rushes because they were up (8) and got the ball back w/< 4min to go. What about the other 56 minutes? That's basic end game strategy for literally any game vs any team. They didnt win solely because of that last drive lol 

It took them a 59 yard pass TD, a 5 yard pass TD, and a 72 yard pass TD to take the lead in the 2nd (outside their 1st FG in which the Chiefs promptly erased after their 1st drive). Defense then held them to a FG the next drive. Tied. Halftime.

Run the ball now? Jacobs 2 yards. Jacobs 0 yards. 3rd & 8. Not ideal. Incomplete. Punt. 1 min 20 secs off the clock. Defense stops KC somehow.

Now run? Incomplete. Complete for 10 yards. Then 3 incompletes. Not great, punt. But they got more than the first drive. 1 min 38 secs off the clock, and would've been more had they not wasted a timeout. Hmmm. Defense stops KC again. 

Now run while still tied? 2 yard Booker run (not a successful play as you know). Incomplete but defensive holding, auto 1st. Hmmmm. Booker 4 yards. Complete 4 yards. Complete 2 yards for a 1st. Timeout KC. Complete 4 yards Jacobs. Complete 23 yards (key play alert). Complete 7 yards. Jacobs 2 yards. Jacobs 0 yards (bleh back to back). Carr 2 yards, converts a 1st down on 4th.

Nooooooow it's the 4th quarter. How many teams are going into the 4th, leading or tied with KC, all while having the ball so they can truly control the clock without anymore stoppage besides the 2 min warning (even though the start of the 4th is still waaaaay too much time to run the game out with but I guess it's theoretically possible)? This year the answer is only LAC (who didnt have the ball to start). KC promptly scored 8 to tie it, then they exchanged FGs to go into OT, but the Chargers get the ball first so that's good! Complete 1 yard. Not a successful play but over the long haul passing still gets more per play. Ekeler 2 yards. Oh. 3 & 7 is not ideal. Probably not smart to run on 2nd & 9 but whatever. Pass complete 6 yards (2 passes for 7 yards is a dreadful average but still better than the run average of 2 on that poss). Punt. KC wins. 

You gotta understand how the Raiders got to there TOP and how they were able to score so much. It's because they're passing was so efficient. Their running actually wasn't efficient...which is I guess what you want? Even though you should probably worry more about scoring on a team that averages 31.1 ppg than grabbing a (slim) lead and hoping your running is bailed out by 23 yard pass plays. Because 4.1 ypc is below the league average of 4.4 and 24% 1st% is not good (Chiefs had the same), especially when compared to their passings 41% (Chiefs had a bad 29%). Even down passing games, they're still converting more. Raiders EPA per rush .08 and for pass .56 but for the Chiefs per rush .06 and their pass .13 hmmm. Success rate? 47 for Raiders rush, 50 for their pass hmm actually pretty close but still passing. 

So yeah you're right. Stress their mediocre defense by attacking...through the air. Running is unlikely to be efficient, over the long haul might I stress again, because it's always been that way. Even when passes are incomplete you can get a holding (like the Raiders did) or DPI call (also like the Raiders got in the 1st q). Not getting many of those on the defense when you run the ball. In fact the opposite. Whenever you see a big run and a yellow flag, who's that usually on? Yeah holding on the offense. That's the nature of running (at least in todays league). 

You get more TOP because of defensive stops. Or rather the opponent offense failing like LACs. And your team not going 3 and out but that happens a lot more when you try to early to "run the clock" aka "establish the run" aka 3 yards and a cloud of dust annnnnd punt. Whatever you wanna call it . Controlling the clock sounds smarter but there's a ton a nuance in how that works. 

Running the clock out just isnt a viable strategy outside the last few minutes (and that goes for any game, like the one and only time you should absolutely choose the low efficient run game, is to force the other team to use timeouts, which the Raiders did, congrats, but I digress). Otherwise the average pass play can also gobble up TOP. Passing doesn't mean no huddle though. But there's more than the last 3:57 (which was also aided by a Tyreek offsides penalty) you gotta worry about. And that other time should generally be a lot of passing. 

You can't stop them from passing. And they can't stop you. You can coax them into running like the Bills did which is the right strategy defensively. They lost but that's because they coaxed themselves into being "balanced" which is on themselves. They could've passed. Buffalo learned from this though. They scrapped that offensive gameplan and when they faced the high flying Seahawks pass attack and decided to pour it on before worrying about TOP (even though they eventually won it BECAUSE they stuck to efficient passing). 

And that's the general league strategy that should be, honestly, employed at almost every game. Offensively attack through the air until the 4th q, hopefully your defense has gotten some stops where you can then run the clock out with a lead and only a few minutes left. Defensively goad them into running and hope you have the pass defense to stop when they don't run.

Now add Atlanta to the list of teams that won the TOP but ended up losing to the Chiefs (Mahomes real lucky that EZ throw wasn't picked). Dont worry about TOP, and just score, until the last few minutes of the 4th. 

37 Chiefs fan, so I have my…

Chiefs fan, so I have my biases.

The parallels with 2011 GB are uncanny.  At the season level, I'm perfectly happy with the Chiefs at #3 or so in DVOA.  They were #4 last year and #1 in 2018.  I'll take the 2019 outcome.  Seriously, I don't have a problem with a model that puts this year's Chiefs as one of the best teams but not a dominant, clearly the best one.  Because I don't think they are dominant or possuming. And obviously they are beatable, they've been beaten.

But while the analysis of 2011 GB vs. 2020 KC is interesting, it's tough to defend the value of DVOA when the #2 26.2% DVOA/ #3 23.4% DVOA in 2011 lost to the #12 7.3%/#14 5.4% team.  Here's the summary in the FO preview of that NYG-GB matchup

The Giants are a good team, they've played very well over the last couple weeks, and they provide particularly difficult matchup issues for the Packers. And yet, it seems almost like people are underestimating the Packers just because they haven't seen Rodgers on the field since December. This was still the best team in the league in 2011. This offense is ridiculously powerful. The Giants are supposed to have a defense that can slow down the Packers and yet the Packers still scored 38 points when these teams first played. This is probably going to be another close game and the Giants will definitely put Aaron Rodgers on his back a few times. Manning will get his points too. But in the end, the Packers are just better, and they are at home. They have to be favored.

As for the DVOA in the KC-NO game.  I get that yards-per-play is a useful metric.  But when one team has 88 plays, 411 yards, and 41:14 TOP vs. 51, 285, and 18:46, along with a higher per-play success rate as was mentioned, I think the difference in yards-per-play is being over-weighted. 

Also, while the fumbled punt return was a huge play, and should be penalized by any descriptive stat, I think it is not predictive when a 3rd-string punt returner makes a bone-headed mistake.  How many times do we think Robinson will field punts over Hardman and Hill in the playoffs*?  At the game-level, a performance measure should penalize the Chiefs for that mistake but as a season-level predictive metric, I'd have to be convinced.

*Hill fumbled a punt in the 2019 playoffs, hell maybe it is predictive.

57 #3 ranking

Heh, as a Colts fan I vastly preferred them "in the top 5" rather than #1 during Manning's hey-day.  I always feel they are more deserving, but acknowledge my bias.  In fact, sitting around 6-8 under the radar is just fine with me.

51 The Giants are a good team,…

The Giants are a good team, they've played very well over the last couple weeks, and they provide particularly difficult matchup issues for the Packers.

This was the key. It was the key for NYG against the Pats, too.

Pretty much every upset came down to matchups. The NYG had a really good d-line, and that overwhelmed GB and NE. SF matched up well with NO -- their D could neutralize NO's running and make them pass (and their pass D was good), but NO's terrible D couldn't stop SF's mediocre offense.

NE escaped Baltimore after they dropped two TD passes in the last minute and then choked a gimme FG to tie. Baltimore was a bad matchup for NE, except for their ST.

The only really weird outcome was NYG beating SF, which wasn't really a favorable matchup for NY. But that was a brutal, ugly game, where SF couldn't convert a 3rd down or field a punt.


The catch is, I'm not really sure which likely playoff teams puts KC into a bind. Theoretically Indy would have the best shot. Maybe the Rams? If they even make the field. Unlike 2011 GB, KC has gone on the road and played pretty much every likely counter-champion, and beaten them.

52 Need the following to beat…

Need the following to beat KC: consistent pressure from front 4, solid corners, and I think most importantly a willingness and the smarts/ability (both from coaches and players) to switch strategies multiple times in the game. If you let Reid and Mahomes get a bead on what you’re doing, you’re toast, no matter how good the personnel.

Miami, LA Rams, Baltimore, and Washington seem most able to hit all of those.

60 I don't think they do, but…

I don't think they do, but like I mentioned last week Washington looks to the defensive players, scheme, and coaching to actually be able to stop a GB offense. Hence me actually being worried about potentially facing them. The formula to stop the KC offense this year is pretty much the same as the formula to stop the GB offense this year. So I can see the WFT def being a concern, but the offense is so bad and the KC def is better than the GB defense (as I've spent a lot of words in this thread on already) that I don't think KC needs to worry that much. They would beat them like GB just did to Carolina who tried to apply the formula. GB having the lesser D, and WFT having better talent than Carolina means WFT would have a chance to score more and GB a chance to score even less than the 24-16 game we say on Saturday.

81 Was mostly talking about…

Was mostly talking about defense's ability to slow down KC offense, not whether the team would win the game. I agree that especially Washington's offense wouldn't be able to keep up even with a slowed down Chiefs' O.



I was considering this when DVOA rated the Chiefs below 0% against the Texans. This was a dominant performance. The yardage was about equal but if you look at the play by play the Chiefs out gained the Texans something like 369 to 360 but 140 of the Texans' yards came in the last 11 minutes of the game when they were down 31 -7 with a 0.1% chance of winning.

This game was a pretty thorough beating. Was it because the Chiefs ran the ball a lot? They were 7 for 13 on third down and 3 for 5 in the Red Zone so it wasn't like they had some freaky success on high leverage downs. Houston was 4 for 10 and both teams were 1 for 1 on 4th. Houston was also 3 for 3 in Red Zone.

And basically the same thing happens against N.O. a better opponent. A closer game. Mahomes way less efficient but still a pretty convincing win to my eye.

As a Chiefs fan will I be worried if they have to face a health Thomas and Brees in the SB? Yes. But DVOA does not take into account the Brees was hurt or Thomas absent and it still rated the Saints as a better team?

There is a serious problem here.

And I generally agree with your take on the Chiefs not being dominant. Only a great team can come back from 24 down but a dominant team never gets 24 down.

106 Respectively disagree. …

Respectively disagree.  There's no serious problem here.  DVOA says that the Chiefs weren't as good as your eyes tell you they were.  That doesn't mean DVOA's wrong, it measures what it measures.  Your eyes, my eyes, everyone else's eyes, those are just our opinions.  And we don't systematically record our "eye tests" every single game, all year long, to develop any sort of rigorous evidence that those eye tests mean a darn thing.   Not that DVOA "means" anything, either.  But at least it's a rigorous attempt to measure the true quality of football teams, at a more fundamental level than "this game was a pretty thorough beating".

112 True Quality

The serious problem is DVOA measures Houston and New Orleans as having more "True Quality" in their losses against KC. In Houston's case that is laughable in New Orleans' it is highly dubious.

66 Quietly watching...

As Josh Allen and the Bills offense beats the teams they are supposed to and the defense gets healthier and plays solid. NO, GB, KC, BUF as top 4 feels right with wc BAL and TB as spoilers. 

70 Worst rush offense DVOA vs NFL average chart

Two things really stick out to me:

1.  How awful running has become in general as NFL DVOA average in 2019 was -8.0% and 2020 is -10.4%  Aaron had mentioned that the difference between the pass and the run has gotten larger in recent years, but this is getting rediculous.

2.  Houston has DeShaun Watson, is anyone really playing to stop the run?

Regarding item (1) I am wondering what an ideal pass run-ratio would be?  Why would I want to stick with a play that is 8% - 10.4% below average as a whole?  Then to expand on that thought, should Houston be throwing almost exclusively? Logic says that teams must be geared up to stop Watson, not the run, yet they do not stop Watson effectively, but do stop the run effectively.

Regarding item (2) above, none of these teams played above .500 except for 11-5, 1986 NE.  Two of these teams went 1-15.  Thus as a whole, the teams on the chart are constantly trailing which would make it easier for them to run as a defense is not going to be concerned about the run when leading by multiple scores late in the game.  For that matter, I would play my defense to encourage all of these teams to run all of the time.

This leads to my last point which is, based on the idea of how to defend these teams, are they even worse at running than the chart would lead us to believe?  In other words, the defense will constantly be looking to defend the pass,  possibly with 5 or 6 DB's and yet these teams still can not run effectively at all.


86 I just remember the comments…

I just remember the comments FROM LAST SUMMER about the Bills being a run-first team and how Diggs would hate it, and now they're having games where the RBs don't have 20 carries, or even touch the ball until the 2nd quarter. They're hardly the only team this applies to.

115 As I type this out it sounds…

As I type this out it sounds kind of like a backhanded compliment, but I mean it sincerely - Allen has to be one of the most improved players of recent history. QBs almost never exceed their passing efficiency from college to the pros - most players achieve a certain completion %, Y/A, etc. in college, take a hit when they enter in the NFL, and (if they pan out) grow back into their old profile. See Bill Connelly's old piece (which places huge red flags on Allen's stats):


Allen has blown his mediocre college efficiency - playing in the Mountain West conference - out of the water in the NFL this season. He was definitely a very high-ceiling talent, but it's still remarkable to me that he appears to be the only QB to pull this off this type of improvement at least the last decade.

118 There is a massive machine…

There is a massive machine that identifies top QB prospects early in middle/high school and directs them into developing their skills, garners the attention of D1 scouts and filters out the unworthy.  The overwhelming majority of top NFL QB prospects have been pushed through this machine since they were 14 or 15 years old, picking up quite a bit of polish on their skills in the process.  Somehow, Allen was missed* by this machine but his raw talent was just otherworldly which is why he attracted Wyoming's attention and later that of the NFL.

For your typical NFL QB prospect, it is probably true that the machine put them somewhere near their peak by their senior season.  For a prospect like Allen (which is rare to begin with), I think the only thing you can say about them coming out of college is "Who the hell knows?".  You could spend 3 years coaching him and get nowhere or (as happened) you get near linear growth and a guy in the MVP conversation.

*Allen grew up not *that* far (maybe 100-150 miles) from Folsom HS which is an absolute powerhouse program that, uh, attracts a lot of Power5 quality talent and regularly competes for the CA state championship.  I've lived in Folsom for 5 years haven't heard anything about him ever being on their radar.  His HS career  must have been absolute stealth mode.


76 Not the Nineteen Nineties

Wait, the 49ers, at 5-9 and fresh off a loss to Dallas, are still in the top half by DVOA? This team has gone 1-3 against the NFC East. Let me repeat: it has a losing record against the NFC East! Sure, injuries, and sure, quarterback issues. But it's amazing to me that a couple big early-season wins are enough to keep them from succumbing to the fearsome DVOA Ratings Page scythe.

The Niners' contribution to Dallas's improbable winning run might also prevent the Cowboys from giving up a long-held title: the team with the best all-time winning percentage in the regular season. The Cowboys overtook Miami some time in the 21st century, but their .5689 record is now mere fractions better than Green Bay's .5687. I can't work out if the Packers need to win two or three games on weeks where the Cowboys lose in order to overtake. I also can't decide what's more impressive: that Dallas still holds this record despite a quarter-century of mediocrity, or that it is likely to be passed by a team that has been in the NFL for a century.

The Packers should at least finally pass Dallas as the team with most seasons in the playoffs. In this instance, existing for an extra 40 years certainly helps. Dallas (33 seasons) has held the record for 25 years, since overtaking that franchise synonymous with postseason play, the Cleveland Browns. (If CLE1 and the Baltimore Ravens are considered the same team, that's your standout by a mile.) Green Bay has already qualified for postseason number 34. I refuse to countenance a scenario where Dallas makes the playoffs in 2020, even though FO's odds say it is feasible. There's been too much strangeness for me to process already this year.

But abysmal as it is, the 2020 NFC East is nowhere near being the worst divison ever by DVOA. (The NFC West of ten years ago looks unbeatable; I'd forgotten how dismal it was.) In fact, it has almost passed me by that none of the worst five teams in 2020 are from the NFC East – or, indeed, from the NFC at all. DVOA thinks this is one of the more lopsided seasons by conference, with NO and TB right up top, and the AFC filling eight of the bottom 12 slots. Does it feel that way? I'm not sure it does, perhaps because a lot of the talk is about the Chiefs (and Steelers) on the one hand, and the pratfalling NFC East on the other. Also, there are no 'average' AFC teams between Tennessee at +9.0% and Cleveland at -4.2%.

88 Aaron is so Smart

He realized the Brady-Manning irrational argument threads were gold, so he sought to replicate it by starting a GB-KC/KC-field/Favre-Rogers thread.

90 The Favre-Rodgers discussion…

The Favre-Rodgers discussion is an interesting one for sure. Packer fans are SO spoiled to have watched these two for the past 28 seasons, yet we also feel so robbed that they only resulted in 2 titles (so far).

Being 34, Favre started playing before I really got into the league, though by his peak seasons of '95 and on is when I started to get into the NFL and the Packers in particular. So a lot of my memories of watching him in his prime are a little muddy now. Plus the fact that I could only get a handful of televised games in those days.

If I had to answer, I would probably say Rodgers is better. However, Favre had a knack for turning losses into wins that I don't as often see from Rodgers. He's not the 4th quarter comeback guy that the media likes to portray. But that's not really a knock on him as much as it is the style his teams have played as front runners. Conversely, Favre could also turn wins into losses at the worst time. Rodgers doesn't really do that. 

Someone here said it best once. If you need one drive, you want Brady or Manning or Brees. If you need ONE PLAY, you want Rodgers.

97 I think DVOA is wrong that…

I think DVOA is wrong that KC outperformed NO. Looking at Quick Reads Mahomes had 78 DYAR & Brees had 1, Bell and CEH each had high 40`s DYAR and no NO RBs showed up. So DYAR would likely tell a different story, more consistent with eyeballs and conventional stats. Conventional stats show KC leading these categories: 1 three and out for KC to 7 for NO, 410 yards to 280, 41 mins TOP to 19, 98 plays to 55, 9 of 18 third down conversions to 1 of 11. These stats do ignore special teams. Sometimes successful drives are more important than successful plays.

104 I agree

KC crushed NO, and a few quirky things that happened kind of disguised that. I keep hearing talking heads say the way to beat KC is to play keepaway, but what if KC possesses the ball for 42 minutes? 

Mahomes is kinda like Jordan or Lebron or David Ortiz in that DVOA has no way to measure how well they will play with the game on the line.

98 1998 Minnesota Vikings DVOA. How did this happen?

Since we are comparing 2011 GB with 2020 KC, we have a theme of 15-1 teams assuming that the Chiefs win out.

Since only one team has gone 16-0, 15-1 is tied for the 2nd best record in the DVOA era.  

The 2004 Pittsburgh Steelers were +121, DVOA 37.6.

The 1985 Bears make sense, their point differential was +258 with a fantastic 52.9 DVOA.  Both of these numbers are among the best in the DVOA era.  Based upon record and point differential they were the equivalent of the 1998 Vikings but.....

With a 27.4 DVOA, the 1998 Vikings hardly qualify as an all time great team in the DVOA era.  They were third in DVOA just in 1998 alone.  How on earth did they go 15-1 with a positive 260 point differential?  Both their record and point differential are among the best in the DVOA era.

The 1998 Vikings played the 24th ranked schedule so that hardly explains it.

To tie the Vikings point differential the Chiefs must win their last two games by a combined 135 points.




102 Vikings

I feel like this has been brought up many times on here over the years. Basically if I remember rightly the Vikings did all of the things well that DVOA doesn't reward: recovering fumbles, gaining 40+ yards on chunk offensive plays, and being good on third down. Their defense was just barely average. They had fewer first downs rushing than their opponents, despite playing with a lead almost all season, and had only 35 more first downs net, which is low for a supposedly elite team.

Plus the 24th ranked schedule is a bit misleading. Already in a league of 30 teams it explains a fair amount. But 1998 was a season with an unusually large range of schedule strength: the best and worst were over/under 10%, which is mighty rare. Division games accounted for more than half of all games in 1998, versus less than 40% now, and when a division was strong or weak, it affected the strength of schedule much more than it has done since 2002. Minnesota's minus-3.5% SoS would have ranked 29th in 2018, for instance.

109 Always take the field

I can't tell if Aaron's "frustration" is with the Chiefs or with how other media talk about the Chiefs. I think the Chiefs are neither great nor dominant.  But they are clearly one of the best teams in the league this season. That may be enough to get them another trophy.

It's unusual for the team with the highest DVOA to win the Super Bowl.  It's also unusual, in the last 20 years at least, for any team not named New England and led by Tom Brady to win a Super Bowl. The Giants, Steelers, and Ravens are the only other teams to have won multiple championships since 2000, having each won twice.

The proper perspective is to acknowledge the flaws on any team under analysis and, when assessing its odds of winning it all, to always bet on the field.

113 It was not unusual for a team with the best DVOA to win a SB

From 1985-1999 ten of the fifteen SB champs had the top DVOA.  In the past twenty seasons only four of the SB Champs had the top DVOA.

If you want to divide it even further, 12 of 19 SB champs had the top DVOA from 1985-2003, in the past 16 years only two SB champs had the top DVOA.  




117 You’d have to be an idiot to…

You’d have to be an idiot to take the Chiefs against the field. The only team in my NFL watching lifetime that comes close to being favourites over the field at the beginning of the playoffs, are the ‘07 Patriots. And the Chiefs are clearly not close to being as dominant as that team. 

119 This just might be a plain,…

This just might be a plain, dumb question but is Variance factored into a teams overall DVOA and/or is it also factored in offensive and defensive DVOAs for their respective variances? 

122 I don't know what model DVOA…

I don't know what model DVOA uses, but I'll try to take a stab.

If you are asking if  "the variance" directly hits DVOA, my guess is probably not. You typically see variance used in static concepts like T stats or Sharpe Ratios.

However, like any predictive model, from regressions to deep learning, estimates vary in their magnitude by how much variance there is in the data. In other words, if your data is very noisy, it will be reflected in it's predictive power

125 It's exactly the opposite…

It's exactly the opposite. Variance is a product of DVOA, not the other way around. It's literally just selecting each team's DVOA in each game and using Excel's VARIANCE function to measure that data set.


120 Variance factored into DVOA or separate?

This just might be a plain, dumb question but is Variance factored into a teams overall DVOA and/or is it also factored in offensive and defensive DVOAs for their respective variances? 

127 Weekly DVOA

I was looking for the weekly DVOA for Week 15 and Week 16, I see that now the link leads to a page which only gives the DVOA as of now. Does anyone know how I can get week specific DVOA? I try to find trends in the team rankings.