The DVOA ratings have finally blown up.

[Rulon Velvet Jones]

http://msn.foxsports.com/nfl/story/6122994

I have nothing to add. When a 2-5 team can be in your top 10, your system is broken and you lose all cred.

[DomCasual]

Uh-oh. I hope Lidderer and yavoon don't see this. You'll be in BIG trouble, mister!

[Kaylore]

It's funny because your DVOA rating doesn't equal wins and the we're the team that proves this the most often. What they had to say about it was pretty funny, though:

Top 10 reasons why DVOA hates the Denver Broncos:
10) Acronym stands for "Denver? Verily, only average."
9) Quarterback has the lowest neck-beard road-rage coefficient in the league.
8) Too much time spent in Coors Field humidor has weakened Broncos DVOA rating.
7) Marilyn Musgrave-Angie Paccione attack ads are poisoning political discourse in this country, and the spreadsheets are very upset about this.
6) FO intern? Roc Alexander.
5) First place rank in rushing touchdowns allowed offset by first place rank in repulsive new art museum wings.
4) Coors Light ads are awful.
3) Local college rub-off effect: Colorado Buffaloes are so bad that they've named Ward Churchill as head coach just to get the team riled up about something.
2) DVOA was designed by Jean-Luc Connerie, a bitter lifelong fan of the Quebec Nordiques.
1) It's a dispassionate statistical system based on studying eight years of NFL play-by-play data, and it recognizes that the Broncos have yet to enjoy a dominating victory and the quarterback was awful for the first seven weeks.

[DomCasual]

I do have to admit that the Coors Light ads are some of the worst ads I have ever seen - just totally stupid. And the Quebec Nordiques thing is kind of funny.

[Taco John]

We're sure to receive an explination that there's nothing wrong with the DVOA, just our understanding of it. It's us who are broken, not the system ranking a 2-5 Pittsburgh team that lost to the (then) 1-5 Oakland Raiders ahead of the 5-2 Denver Broncos.

[Mediator12]

DVOA is pretty accurate on the whole as an evaluator of past team performance. The Broncos have not been dominant and have not put teams away when they have had the chance. I think people forget they do a good Job of incorporating ST's and Field position into the formula when evaluating performance.

They do not look at records, which most fans always like to point at the Scoreboard, but the potential to succeed in the future based on past performance. They really are like stock analysts who are trying to see what is likely to happen, not what exactly is going to happen.

When you see that PIT is ranked higher than DEN it only means that they are more likely to make more plays in the game than Denver does. That is pretty accurate. What it can not account for is the hot and cold play of a QB who has had a 140+ rating against ATL and then next week goes out and has a 40+ performance against the Raiders with two INT's returned for TD's. It is a probability thing that says that PIT should win 5 games if they played OAK 6 times. You never know when that game the Raiders will win will occur.

It is a lot more accurate than the what have you done for me lately overreacting power Rankings out there.

[UKBronco]

I don't entirely understand whether the DVOA thing is based on some statistical thing, but I found the article fairly amusing. More suited to Page 2 perhaps.

Definately not something to get riled about

[DomCasual]

We're sure to receive an explination that there's nothing wrong with the DVOA, just our understanding of it. It's us who are broken, not the system ranking a 2-5 Pittsburgh team that lost to the (then) 1-5 Oakland Raiders ahead of the 5-2 Denver Broncos.

Simpleton.

[Mediator12]

DVOA is a completely objective measure of past performance by every team. It breaks down success play by play and does not look at W-L per se, but yardage and First down correlations to points.

The reason Denver's defense had not been so high in the rankings is because they have had a higher yards per drive and FD's allowed than say JAX, BAL, CHI and the like. Those extra opportunities constrict the opportunities of the offense to have good FP and therefore score points as well.

[Taco John]

DVOA is a completely objective measure of past performance by every team. It breaks down success play by play and does not look at W-L per se, but yardage and First down correlations to points.

The reason Denver's defense had not been so high in the rankings is because they have had a higher yards per drive and FD's allowed than say JAX, BAL, CHI and the like. Those extra opportunities constrict the opportunities of the offense to have good FP and therefore score points as well.



It doesn't mean anything when 3 of the 5 teams we beat are rated ahead of us. It's just some arbitrary scheme someone is using to put together clever articles and get paid by Fox. Good for Schatz for swinging a paycheck out of it. That's about all it's good for.

[Mediator12]

It doesn't mean anything when 3 of the 5 teams we beat are rated ahead of us. It's just some arbitrary scheme someone is using to put together clever articles and get paid by Fox. Good for Schatz for swinging a paycheck out of it. That's about all it's good for.

That is simply not true. What you are missing in Translation is that each team is theoretical in their performance against the league average. It also adjusts for strength of schedule. It does not compare who has played who, that is a process called beatpaths that most fans think has no flaws when in fact it is full of them.

DVOA can never tell you who is going to match up better, execute better, have a better gameplan, and when fluke's like two defensive TD's are good enough to beat ARI after they were leading by 20 points in the second half. That is why they play the games.

Aaron Shatz and I have had words in the past and I do not think they are perfect. However, what they do is very accurate. It is probability based, not opinion based. They have the best record of predicting the playoff teams preseason the last three years running using this system. Yeah, that is great and all, but they have merit in finding performance issues and areas that need improvement.

Come on TJ, they think Plummer Sucks unless he is running ;D That should at least get you to read their explanations.

[Rock Chalk]

The problem with probability based stats is that they do not take in that mystical entropic quality that occurs in probability.

Furthermore, more than once DVOA has altered their rankings contrary to what their statistics say. They, at one time, had Indy ranked lower than 4 when they were unbeaten and had been destroying folks but had Denver ranked number 1 (this was either last yer or the year before). They arbitrarily placed Indy on top because, like everyone and their mother who actually WATCHED the games, KNEW Indy was the best team AT THAT TIME.

They had to adjust their rankings regardless of what DVOA told them based on pure common sense.

DVOA is fun, but is inherently flawed. Its probability predictions are missing about 300 variables for any true accuracy. That being said, it is about as accurate as any kneejerk "ranking" system in that by the end of the year its fairly good at predicting who will be in teh playoffs. IT TAKES THE WHOLE YEAR FOR IT TO BE TRULY ACCURATE. And by then, any moron with a brain (including most of the dip****s on this site) can figure it out.

[Taco John]

That is simply not true. What you are missing in Translation is that each team is theoretical in their performance against the league average.

I understand that. What I don't understand is why that matters more than, you know -- actually meeting on the field.

Come on TJ, they think Plummer Sucks unless he is running ;D That should at least get you to read their explanations.



Now, I'm not saying we should throw the baby out with the bathwater here... :~ohyah!:

[watermock]

Silly us Med, for thinking wins and losses are what count. I guess we are all simpletons. Stats are for baseball, football uses a ball that bounces with a mind of it's own.

[Taco John]

I just want to point out that ratings like this are why the NFL is, and always will be, superior to the college game. In the NFL, these useless ratings charts have no bearing on anything except for who gets paid to write about it. In college, they determine the champions.

God bless the NFL.

[Bob's your Information Minister]

Bennett just got healthy the last two weeks. He did a nice job against the Seahawks. I expect he'll continue getting 5-10 carries a week from here on out.

Also, Vic Carucci had KC in his top ten. :D

http://nfl.com/news/story/9767940

9. Kansas City (4-3; unranked): With Larry Johnson (http://nfl.com/players/playerpage/396164) rolling and Damon Huard (http://nfl.com/players/playerpage/1748) filling in superbly at quarterback, the Chiefs have emerged as a force in what suddenly has become one of the league's most competitive divisions.

[El Guapo]

check out the comment for the Cowboys:


At this point, Jason Witten has taken over for Jeb Putzier as "fantasy tight end who teases you with yardage but never seems to get a look in the end zone."


:rofl:

[Atwater His Ass]

I'd rather have a DVOA system that draws a line between stats and opinon. I enjoy reading the DVOA rankings much more than say, Dr. Z's bull**** gut feelings on where he or anyone else thinks the teams rank.

And really, it doesn't matter. Win and losses matter and making the playoffs and winning those games matter. I could care less if Indy, Pitt, NE, Baltimore, SD and Jax are rated ahead of us in DVOA. I just want to beat those teams in the playoffs, where it matters.

[Mediator12]

Silly us Med, for thinking wins and losses are what count. I guess we are all simpletons. Stats are for baseball, football uses a ball that bounces with a mind of it's own.

Wins and losses are what matters, mock. REPEAT, wins and losses are what matters.

This is a method to assess future performance and rank teams based on past perfromance. It is flawed, but it is also one of the best systems out there at predicting future outcomes and teams strengths and weaknesses.

The DL statisitcs say that Denver has an average run defense and last season the least effective pass rush, would you disagree with that? This season they are only slightly improved. It says that even though Denver is 5-2, the QB has been well below average, would you disagree with that? It says that even though the defense has been lights out in points allowed, until last week, they have been only slightly above average in allowing Field position.

[listopencil]

The DL statisitcs say that Denver has an average run defense and last season the least effective pass rush, would you disagree with that? This season they are only slightly improved. It says that even though Denver is 5-2, the QB has been well below average, would you disagree with that? It says that even though the defense has been lights out in points allowed, until last week, they have been only slightly above average in allowing Field position.

It also says that two teams we beat are better than us, would you disagree with that? I would. I think it's crap.

[ScottXray]

Wins and losses are what matters, mock. REPEAT, wins and losses are what matters.

This is a method to assess future performance and rank teams based on past perfromance. It is flawed, but it is also one of the best systems out there at predicting future outcomes and teams strengths and weaknesses.

The DL statisitcs say that Denver has an average run defense and last season the least effective pass rush, would you disagree with that? This season they are only slightly improved. It says that even though Denver is 5-2, the QB has been well below average, would you disagree with that? It says that even though the defense has been lights out in points allowed, until last week, they have been only slightly above average in allowing Field position.

Using DVOA positions as general predictions then, we should go like this:

At Pitt...LOSS
At Oakland WiN
San Diego...LOSS
at KC....LOSS
Seattle...WIN
At SD.....LOSS
At ARIZ....WIN
BENGALS...WIN
49ers....WIN
Leaving us at 10-6 and MAYBE going to Indy for final annihilation, or more likely, at Baltimore in the opening round. If we survive THAT we get to go to INDY and help the Peyton Manning legend grow.

In order for this to happen, the defense has to collapse, as it has done before at mid season. I don't think THIS defense will.

Guess we'll see what happens.