Von Miller: The Myth of Double-Teams

[Fantaztic7]

To be clear...if this thread was started with a conversation about 2019 week 1 I would be typing different posts. There are things I saw in week 1 I did not like. Not sure what is up about the things I saw at this time but it had nothing to do with double teams or lack of double teams.

It was obvious from post 1 this was just as much about 2018 as 2019....who knows what was thought about more when making this thread.

In one of the threads following Monday night’s game there was a comment about Von being double-teamed. It’s not difficult to get the context of this thread, given how quickly you posted the same Next Gen stats from last season.

The common defense Von doesn’t show up has been and continues to be something like, “He’s always double-teamed.” You can parse the discussion by challenging the meaning of “constant” and “consistently”. Again, it’s not hard to interpret the main theme - some fans defend Von with a generic comment about being double-teamed.

Monday night’s game was no exception with the same excuse for Von. The fact is that he was only double-teamed twice (from the tape I watched). The lack of pressure wasn’t only a result of Oakland’s scheme with quick passes and their talented RT. Von was slow with his first step. He gave minimal effort IMO and looked disinterested based on the lack of intensity. He wasn’t a factor defending the run and the FB absolutely smacked him on one play.

As far as the Next Gen stats, do you know how they evaluated the double-teams? Was the stat based on pass plays, run or a combination? Do you know the accuracy of those stats? You base your entire argument on those stats so please provide some background on how it’s measured.

[Fantaztic7]

I answered in another post but wanted to show you the respect to answer directly

No I do not. I researched for about 30 mins because I wanted a deeper look.

Like others claimed I also watched game film but many times I have seen different posters see different things watching the same game film. I referenced that game in my post as well as some examples. I am not going to discuss it at this point. One of the reasons is because the examples from that game I used tactfully to help me determine if people had an open mind or not.

If people still have feelings about that thread this many months later I am not sure if these forums can handle that discussion.

Mods have enough work to do. Let us not add to that workload.

I missed that post, my oversight and thanks for answering.

John Turney of Pro Football Journal researched the same question and asked Next Gen how they analyze the film to generate the stat. Next Gen would only say the stat reflects double-teams on pass plays. Turney did the analysis of the film and found 70% for Aaron Donald was misleading. Turney watched every play and determined at least 18% of the double-teams never occurred. The percentage of double-teams was inflated.

I can’t trust Next Gen’s stat on double teams when they won’t even define how they measure it.

The fact is that Von isn’t double-teamed as much as some fans believe. It’s a reflexive excuse when Von plays poorly.

[Hadez]

In one of the threads following Monday night’s game there was a comment about Von being double-teamed. It’s not difficult to get the context of this thread, given how quickly you posted the same Next Gen stats from last season.

The common defense Von doesn’t show up has been and continues to be something like, “He’s always double-teamed.” You can parse the discussion by challenging the meaning of “constant” and “consistently”. Again, it’s not hard to interpret the main theme - some fans defend Von with a generic comment about being double-teamed.

Monday night’s game was no exception with the same excuse for Von. The fact is that he was only double-teamed twice (from the tape I watched). The lack of pressure wasn’t only a result of Oakland’s scheme with quick passes and their talented RT. Von was slow with his first step. He gave minimal effort IMO and looked disinterested based on the lack of intensity. He wasn’t a factor defending the run and the FB absolutely smacked him on one play.

As far as the Next Gen stats, do you know how they evaluated the double-teams? Was the stat based on pass plays, run or a combination? Do you know the accuracy of those stats? You base your entire argument on those stats so please provide some background on how it’s measured.

Yes I agree Von had issues in week 1 against the Raiders. What I saw made no sense compared to anything I have seen up until now with Von so I am waiting another week of data before commenting. We do not know the week to week assignments or play calls. I usually look for multiple examples of something before forming an opinion....even more so before expressing it.

[FlaccoTruther]

In another thread there was a comment, the often repeated defense of Von Miller on being double-teamed. Another forum member successfully disproved the double-team myth last season.

Von Miller wasn’t consistently double-teamed and chipped vs Raiders on Monday night. In pass protection the Raiders only double-teamed Von twice. Looking at the tape, a defensive coordinator should no longer plan help against Von (unless they have a really bad RT). The TE had an easy time blocking Von one on one.

Even without pass plays designed for quick release, it didn’t look like Von would have been a factor. His first step looked slow and effort was weak. It could be the lack of reps in preseason, but he looks out of shape and/or disinterested. He was smacked by the fullback on one play.

We’ve seen stretches of games where Von hasn’t shown up. He looked and played less than elite on Monday night. He was quite mediocre if not below average as an edge rusher. He also looked fairly weak on run defense.

Anyway, the main point is that the constant double-teams are a myth.

I'm not buying that narrative. He was given extra attention and Carr was getting the ball out in 2.33 seconds. If your CB's can't cover and allow the QB to get it out that quickly nothing really matters from a pass rush perspective. A lot of that was CB's playing too safe and giving their WR's too much of a cushion I don't know if that was scheme or the CB's but I suspect it to change next week.

I'm also not as scared of the Bears offense to be honest. They have nothing. It's turnovers on offense and finding a way to score that worries me. With all that I said earlier they did hold Oakland to 24 points which was league average last week (despite the TOP being extremely lopsided) so they still did enough things well to give the Bronocs the a chance to win. More then the defense it's the offense that needs to be the focus going forward.

16 points just isn't going to cut it

[Hadez]

I missed that post, my oversight and thanks for answering.

John Turney of Pro Football Journal researched the same question and asked Next Gen stats for how they analyze the film to generate the stat. Next Gen would only say the stat reflects double-teams on pass plays. Turney did the analysis of the tape and found 70% for Aaron Donald was misleading. Turney watched every play and determined at least 18% of the double-teams never occurred. The percentage of double-teams was inflated.

I can’t trust Next Gen’s stat on double teams when they won’t even define how they measure it.

The fact is that Von isn’t double-teamed as much as some fans believe. It’s a reflexive excuse when Von plays poorly.

I understand not trusting when not knowing the methodology. There is even more to it than that.

For example the game I watched there were plays when the offense called for multiple blockers on Von but for whatever reason the multiple blockers did not touch Von. Does a stat gatherer count those plays? I mean Von technically got beat by less than a double/triple team. The offense may have planned for it and an offensive guy may have done nothing but stand there and watch Von waiting for Von to make an athletic "Matrix" move but technically he never touches him...what does it count as?

There are some plays I think are obvious...and yet people see different things.

There are other plays in which they are clearly vague and I can totally understand calling them either way.

Still a clear definition of what plays are counted and which plays are not counted helps us understand the baseline of the numbers we are discussing.

edit =

In this case it also matters what are we talking about. If we are talking about Von's ability then in my example above I do not think we count it because Von still got beat without the other guy touching him. If we are talking about attention from the offense than we count it because the offense planned for it and the offensive player did nothing but stand there.

Sorry if I got too deep. I work for avionic corporate manufacturing and analyzing metrics to further business goals is what I do.

[Fantaztic7]

Yes I agree Von had issues in week 1 against the Raiders. What I saw made no sense compared to anything I have seen up until now with Von so I am waiting another week of data before commenting. We do not know the week to week assignments or play calls. I usually look for multiple examples of something before forming an opinion....even more so before expressing it.

It made sense to see the same lack of effort because it happened in multiple games last season (and others saw the same thing). Vic Fangio’s comment about Von being able to improve wasn’t by accident.

[Fantaztic7]

I understand not trusting when not knowing the methodology. There is even more to it than that.

For example the game I watched there were plays when the offense called for multiple blockers on Von but for whatever reason the multiple blockers did not touch Von. Does a stat gatherer count those plays? I mean Von technically got beat by less than a double/triple team. The offense may have planned for it and an offensive guy may have done nothing but stand there and watch Von waiting for Von to make an athletic "Matrix" move but technically he never touches him...what does it count as?

There are some plays I think are obvious...and yet people see different things.

There are other plays in which they are clearly vague and I can totally understand calling them either way.

Still a clear definition of what plays are counted and which plays are not counted helps us understand the baseline of the numbers we are discussing.

There are different ways to view a play, the example of the holding penalty on Bolles in preseason illustrated the point (TE missed the chip).

At the same time I believe Von gets a pass for a poor games with a generic reflex on double-teams.

The analysis by johnlimburg was correct based on the tape from multiple games. Unless someone breaks down the same games and counters with substantial evidence, the conclusion stands.

[Fantaztic7]

I understand not trusting when not knowing the methodology. There is even more to it than that.

For example the game I watched there were plays when the offense called for multiple blockers on Von but for whatever reason the multiple blockers did not touch Von. Does a stat gatherer count those plays? I mean Von technically got beat by less than a double/triple team. The offense may have planned for it and an offensive guy may have done nothing but stand there and watch Von waiting for Von to make an athletic "Matrix" move but technically he never touches him...what does it count as?

There are some plays I think are obvious...and yet people see different things.

There are other plays in which they are clearly vague and I can totally understand calling them either way.

Still a clear definition of what plays are counted and which plays are not counted helps us understand the baseline of the numbers we are discussing.

edit =

In this case it also matters what are we talking about. If we are talking about Von's ability then in my example above I do not think we count it because Von still got beat without the other guy touching him. If we are talking about attention from the offense than we count it because the offense planned for it and the offensive player did nothing but stand there.

Sorry if I got too deep. I work for avionic corporate manufacturing and analyzing metrics to further business goals is what I do.

No apology necessary as it’s part of your thought process.

[Hadez]

There are different ways to view a play, the example of the holding penalty on Bolles in preseason illustrated the point (TE missed the chip).

At the same time I believe Von gets a pass for a poor games with a generic reflex on double-teams.

The analysis by johnlimburg was correct based on the tape from multiple games. Unless someone breaks down the same games and counters with substantial evidence, the conclusion stands.

That is your conclusion and the conclusion of other(s). In science they call it a theory and allow others to test the theory.

Based on the fact the two people I am talking with neither remembered the game I said I watched nor remember the example I used I would say it is not a very well tested conclusion.

If there would have been some critical thinking based discussion based on the example I presented which allowed me to understand what baseline the discussion was talking about since offensive attention was mentioned while it was clear I think Von performance was in question maybe I would have spent more time watching more games or given more examples of the game I watched.

My conclusion was minds were made up and nothing I posted was going to change anything so why even bother :brick:

[Fantaztic7]

That is your conclusion and the conclusion of other(s). In science they call it a theory and allow others to test the theory.

Based on the fact the two people I am talking with neither remembered the game I said I watched nor remember the example I used I would say it is not a very well tested conclusion.

If there would have been some critical thinking based discussion based on the example I presented which allowed me to understand what baseline the discussion was talking about since offensive attention was mentioned while it was clear I think Von performance was in question maybe I would have spent more time watching more games or given more examples of the game I watched.

My conclusion was minds were made up and nothing I posted was going to change anything so why even bother :brick:

Re-posting Next Gen stats isn’t critical thinking (inaccurate stats by the way).

[Fantaztic7]

I'm not buying that narrative. He was given extra attention and Carr was getting the ball out in 2.33 seconds. If your CB's can't cover and allow the QB to get it out that quickly nothing really matters from a pass rush perspective. A lot of that was CB's playing too safe and giving their WR's too much of a cushion I don't know if that was scheme or the CB's but I suspect it to change next week.

I'm also not as scared of the Bears offense to be honest. They have nothing. It's turnovers on offense and finding a way to score that worries me. With all that I said earlier they did hold Oakland to 24 points which was league average last week (despite the TOP being extremely lopsided) so they still did enough things well to give the Bronocs the a chance to win. More then the defense it's the offense that needs to be the focus going forward.

16 points just isn't going to cut it

How much extra attention?

[Hadez]

Re-posting Next Gen stats isn’t critical thinking (inaccurate stats by the way).

Neither is being so focused on the point that one is trying to make that one is not paying attention to another poster enough to take into their claims of watching game film or their example given with qtr and time on the clock as to the plays they watched.

[Hadez]

Re-posting Next Gen stats isn’t critical thinking (inaccurate stats by the way).

Would you mind linking where the stats where found to be inaccurate?

[Fantaztic7]

Would you mind linking where the stats where found to be inaccurate?

http://nflfootballjournal.blogspot.com/2018/11/aaron-donald-double-teamed-70-of-time.html

Thursday, November 1, 2018
Aaron Donald Double-teamed 70% of the Time on Pass Plays?
OPINION
By John Turney

“While cruising through Twitter I came across this Tweet that stats that according to the NFL Next Gen Stats superstar defensive tackle Aaron Donald be being double team blocked 70% of the time—

After a couple of clarification Tweets with Mr. Long it turns out it was 70% of the time on passing plays. It seemed high to me, having a little bit of experience with these kinds of esoteric stats over the years and also doing some film grading myself and mentioned similar things HERE about Pro Football Focus's research.

So, I pulled up the All-22 film from NFL Game Pass and watched all the defensive snaps of Donald on passing downs. In my view, I was very generous in grading what plays were double teams. If an offensive lineman or back was even assigned to Donald but never really engaged I gave that a "double team" grade. If there was a tackle-end game or end-tackle game and two players, over the course of the snap engaged, I counted that.

All said I came up with a very high figure, an amazing figure in my view, of 52% of the passing snaps Donald was doubled. Fifty-two percent!. Having some knowledge in this, that is as high as it can get for a season. You can find games where a player is doubled more, but not many, if any seasons.

So, what about the Nex Gen Stats figure of 70%?

I cannot say that I am right and they are wrong. I just wish we could find out the methodology and how the 70% number was arrived at. Certainly, I am fallible and could have made errors but 18% off? I don't think so. So, while I won't say I am right and they are wrong but by the same token, I am not going to just accept that I am wrong and they are right.

So, here is one person asking if the Next Gen folks were grading the All-22 or if they were using the chips and computer to do for this stat like they did for the Fastest NFL players or Fastest Sacks or Hurries something like "real-time location data, speed, and acceleration for every player, every play on every inch of the field".

I'd like to know if this was something done visually or on a computer screen. Not that anything would be wrong, but I'd like to know because I think I can see a double team and as mentioned, grading very generously my view is Donald is double teamed on pass players around 52%.”

[Fantaztic7]

Neither is being so focused on the point that one is trying to make that one is not paying attention to another poster enough to take into their claims of watching game film or their example given with qtr and time on the clock as to the plays they watched.

I acknowledged the post and purposely didn’t respond to the comments/questions about the game. The analysis didn’t provide any data, rather it offered questions about what counts as a double team vs what doesn’t. It’s fine to disagree on what constituted a double team here or there. If you’ve broken down the film, provide the number of times he was double teamed in your opinion.

You’ve been doing home projects - same here which is a nice coincidence! If you’re tired from the day no worry about breaking the numbers down from that game.