Picking up Edmunds Option a Rare Beane Mistake

[BADOLBILZ]

1 hour ago, Coach Tuesday said:

 

The above ^^ is what YOU SAID.

 

 

Now you're moving the goalpost.

 

Why don't you just admit you didn't watch the clip being discussed?  Because if you had you would never have called it an example of "a MLB blanketing his coverage like he was a Tre White clone."  HE LITERALLY TAKES TWO STEPS TO HIS LEFT TO DISRUPT A PASSING LANE.

 

Please.  There are reasonable bases to defend Edmunds' play, and then there's hyperbole and outright mythology.  

 

 

Now Edmunds has elevated his game to being a Tre White clone too.   😆

 

We'll be looking for that Tre White "Peanut Punch" next.............oh wait...........no that's right,  fumble recoveries are totally random........no correlation between a well-timed, intentional stripping away of the football from an offensive player fighting for extra yardage......and the subsequent alert defender anticipating the possibility of such.......must be a highly unexpected and random occurrence. 

 

Funny how the same guys keep making all the random big plays...........as I was watching that Milano recovery I was reminded of the Poyer/Milano strip/recovery that was overturned at the very end of the Colts playoff game.    I had no idea this stuff was totally random.

 

 

[BADOLBILZ]

1 hour ago, Alphadawg7 said:

You clearly missed the sarcasm in using the same extremities that I was countering.  
 

And he executed that play perfectly, and NOT all LBs do that, Beasley makes LBs look silly on the same route regularly.  And he did this on more than just this play.  
 

Sorry, you’re still downplaying good plays for the purpose of exaggerating bad plays to confirm your opinion of him.

 

There isn’t a GM in football who wouldn’t have picked up Edmunds option just like Beane did.  
 

This is IMHO the worst thread of the year full of over exaggerated hot takes and non sense.  I don’t know if I’ve ever seen a thread with more blatant confirmation bias attempts as this one.  

 

 

You are confusing "perfect" execution with the minimally successful execution.

 

Perfect execution is baiting Henicke into throwing into the double team and making a play on the football............getting him to move to his second read and putting the responsibility on Milano to handle the much tougher coverage assignment was the least that Edmunds should have accomplished.

 

Again.........Milano had the much harder job and Poyer made an alert and instintive play.     All Edmunds did was combine on a double team to take away a very short pass option.

 

As for the "Beasley makes LB's look silly" take.........sometimes.........but other times he gets bracketed and Allen should be moving on to his next read.    Heinicke did a poor job of recognizing the coverage pre-snap.   Where else was Edmunds going to go if he didn't blitz?  

[Alphadawg7]

21 minutes ago, BADOLBILZ said:

 

 

You are confusing "perfect" execution with the minimally successful execution.

 

Perfect execution is baiting Henicke into throwing into the double team and making a play on the football............getting him to move to his second read and putting the responsibility on Milano to handle the much tougher coverage assignment was the least that Edmunds should have accomplished.

 

Again.........Milano had the much harder job and Poyer made an alert and instintive play.     All Edmunds did was combine on a double team to take away a very short pass option.

 

As for the "Beasley makes LB's look silly" take.........sometimes.........but other times he gets bracketed and Allen should be moving on to his next read.    Heinicke did a poor job of recognizing the coverage pre-snap.   Where else was Edmunds going to go if he didn't blitz?  

 

Come on...baiting him now is the prerequisite, and on a play where Heineke DID throw an interception forcing the pass when no one was open?  You just keep moving the goal post on Edmunds to down play when he does something that is counter intuitive to your "he sucks" agenda.  I mean you comments here are amongst the worse takes so far.  

[GoBills808]

40 minutes ago, BADOLBILZ said:

 

We'll be looking for that Tre White "Peanut Punch" next.............oh wait...........no that's right,  fumble recoveries are totally random........no correlation between a well-timed, intentional stripping away of the football from an offensive player fighting for extra yardage......and the subsequent alert defender anticipating the possibility of such.......must be a highly unexpected and random occurrence. 

 

Funny how the same guys keep making all the random big plays...........as I was watching that Milano recovery I was reminded of the Poyer/Milano strip/recovery that was overturned at the very end of the Colts playoff game.    I had no idea this stuff was totally random.

 

 

To that end (and I realize it's getting off-topic but still relevant vis a vis a player's skillset, real or imagined): this has been studied by several outlets, FootballOutsiders as referenced before as well as FootballPerspective and others. It may seem counterintuitive to you but there is no predictive data to be derived from fumble recoveries. It is entirely random and using them as proof of a player's relative abilities isn't going to offer anything in the way of insight. It's a totally useless endeavor.

 

If fumble recoveries were a skill and not luck, we would expect a nonrandom set of values of a linear regression between the relationship of a team's fumble recoveries from the first half of the season to the second. Instead, you get this:

 

...ie a totally random distribution. The relationship between a team's fumble recoveries from the first half of the season to the second is <.02, or a  nonexplanatory independent variable. Contrast this to the observed values of a team's yards per carry, which IS predictive over time, distributed over the same first half/second half of the season model:

 

 

...you get a very neat linear progression. tldr fumble recoveries are known quantities , entirely random, and should not be used to infer the quality (or lack thereof) of a player/defense/team

[NewEra]

2 minutes ago, GoBills808 said:

To that end (and I realize it's getting off-topic but still relevant vis a vis a player's skillset, real or imagined): this has been studied by several outlets, FootballOutsiders as referenced before as well as FootballPerspective and others. It may seem counterintuitive to you but there is no predictive data to be derived from fumble recoveries. It is entirely random and using them as proof of a player's relative abilities isn't going to offer anything in the way of insight. It's a totally useless endeavor.

 

If fumble recoveries were a skill and not luck, we would expect a nonrandom set of values of a linear regression between the relationship of a team's fumble recoveries from the first half of the season to the second. Instead, you get this:

 

...ie a totally random distribution. The relationship between a team's fumble recoveries from the first half of the season to the second is <.02, or a  nonexplanatory independent variable. Contrast this to the observed values of a team's yards per carry, which IS predictive over time, distributed over the same first half/second half of the season model:

 

 

...you get a very neat linear progression. tldr fumble recoveries are known quantities , entirely random, and should not be used to infer the quality (or lack thereof) of a player/defense/team

You have eyes.  Watch Edmunds closely when the ball is on the ground.  Most of the time he doesn’t react.  When he does, he flops around on the ground like a fish trying to get control of it.  Hand- eye coordination isn’t luck or random.  Some athletes are better than others at seeing, reacting and securing a football that is rolling around on the ground.  Edmunds has zero recoveries.  Milano, who is always hurt, has 7 playing far fewer snaps.  There are random aspects involved in recovering a fumble, but that doesn’t mean that some players are better at it than others.  It’s not 💯 lucky.  There’s some skill involved. 

[GoBills808]

7 minutes ago, NewEra said:

You have eyes.  Watch Edmunds closely when the ball is on the ground.  Most of the time he doesn’t react.  When he does, he flops around on the ground like a fish trying to get control of it.  Hand- eye coordination isn’t luck or random.  Some athletes are better than others at seeing, reacting and securing a football that is rolling around on the ground.  Edmunds has zero recoveries.  Milano, who is always hurt, has 7 playing far fewer snaps.  There are random aspects involved in recovering a fumble, but that doesn’t mean that some players are better at it than others.  It’s not 💯 lucky.  There’s some skill involved. 

The data show otherwise.

[eball]

7 minutes ago, NewEra said:

You have eyes.  Watch Edmunds closely when the ball is on the ground.  Most of the time he doesn’t react.  When he does, he flops around on the ground like a fish trying to get control of it.  Hand- eye coordination isn’t luck or random.  Some athletes are better than others at seeing, reacting and securing a football that is rolling around on the ground.  Edmunds has zero recoveries.  Milano, who is always hurt, has 7 playing far fewer snaps.  There are random aspects involved in recovering a fumble, but that doesn’t mean that some players are better at it than others.  It’s not 💯 lucky.  There’s some skill involved. 

 

I’ve come to the conclusion that Edmunds is a pretty good LB who works well within the Buffalo defense.  Is that the guy I want to pay top-5 LB money to when his contract expires?  Nope.  But he’s also not JAG.  The retention of Milano last offseason looks more and more like the steal of the offseason.  If they both stay healthy the Bills’ D will be just fine.

 

[97bills]

5 hours ago, FilthyBeast said:

Maybe it's just me but it seems that all the big plays given up yesterday (specifically during the WTF mini run in the 2nd quarter) that Edmunds seemed to be a liability or missed a tackle.

 

I still will never understand what the coaching staff sees in him but there's really nothing that stands out about his game and even when he has had opportunities to make game changing plays (i.e. INT) it just doesn't happen.

Remember his still young,oh wait his been in the league 4 years now I guess we can’t use that anymore.  I for one think his just a ok player and would love for us to get better at the position. 

[YattaOkasan]

3 hours ago, Rochesterfan said:

 

2 hours ago, NewEra said:

Agreed 💯.  
 

But he kinda stinks sometimes.  Right?  🤪

Certainly got some things to work on. Again I have no clue on his valuation but the point I want to make is that he’s doing things well that are not obvious to us fans. 

2 hours ago, Fan in Chicago said:

An opinion (mine as well) that Beane should not have picked up his 5th year option is not the same as the others on your list. He is an acceptable player but we can do better. McBeane should have waited to see his performance this season before making decisions about his future with the team. If he left in FA next offseason, so be it. Draft or look for options in FA to get, at worst, similar performance. 

This sorta doesn’t make sense. If you don’t pick up his 5th year you have to make a decision about his future after this year. By picking his 5th year up you get an extra year to evaluate.  You don’t pick up the 5th year if you’ve seen enough and you don’t want to re-sign (sounds like where you are at). Some do re-sign but I think the majority don’t.  Beane did NOT make any long term decisions about his future. He just gave himself more time to evaluate. 
 

the draft is super unlikely to give you similar performance and I also think a UFA would be pretty pricey (12 mil probably) for a top 10 mlb. 
 

I just don’t see them moving on from a three year captain, 2x pro bowler, ranked in top 10 at position by peers when they could get an extra year of team control. 
 

in conclusion yes it’s pretty crazy to call picking up the 5th year a mistake. It’s probably the least crazy thing on my list but sounds like you believe some of the other items on my list. 

Edited September 27, 2021 by YattaOkasan

[GunnerBill]

5 hours ago, GoBills808 said:

No, sorry. Fumble recoveries are not predictive and have zero correlation year to year.

 

Recovery of a fumble, despite being the product of hard work, is almost entirely random.

 

Stripping the ball is a skill. Holding onto the ball is a skill. Pouncing on the ball as it is bouncing all over the place is not a skill. There is no correlation whatsoever between the percentage of fumbles recovered by a team in one year and the percentage they recover in the next year. The odds of recovery are based solely on the type of play involved, not the teams or any of their players.

Fans like to insist that specific coaches can teach their teams to recover more fumbles by swarming to the ball. Chicago's Lovie Smith, in particular, is supposed to have this ability. However, in Smith’s first three seasons as head coach of the Bears, their rate of fumble recovery on defense went from a league-best 76 percent in 2004 to a league-worst 33 percent in 2005, then back to 67 percent in 2006.

Fumble recovery is equally erratic on offense. In 2008, the Bears fumbled 12 times on offense and recovered only three of them. In 2009, the Bears fumbled 18 times on offense, but recovered 13 of them.

Fumble recovery is a major reason why the general public overestimates or underestimates certain teams. Fumbles are huge, turning-point plays that dramatically impact wins and losses in the past, while fumble recovery percentage says absolutely nothing about a team's chances of winning games in the future. With this in mind, Football Outsiders stats treat all fumbles as equal, penalizing them based on the likelihood of each type of fumble (run, pass, sack, etc.) being recovered by the defense.

Other plays that qualify as "non-predictive events" include blocked kicks and touchdowns during turnover returns. These plays are not "lucky," per se, but they have no value whatsoever for predicting future performance.

 

 

The famous Jim Schwartz study when he worked as an intern for Belichick in Cleveland actually went further and concluded that fumbles in their entirety are entirely random and equally non-predictive year on year.

[NewEra]

21 minutes ago, GoBills808 said:

The data show otherwise.

And common sense tells you that people can actually be better at such a thing than others.  Being near the fumbled ball is random.  Going through the process of locating the ball, picking up the ball, fighting for the ball and securing is not.  Tremaine has shown that he is deficient in that department. All the data I need is my eyes.  Turn on the tape and watch what he does when the ball is fumbled.  Your data doesn’t prove that one athlete can’t be better at recovering fumbles than another. 
 

tremaine has had opportunities…..he just sucks at cashing in on those opportunities. 

28 minutes ago, eball said:

 

I’ve come to the conclusion that Edmunds is a pretty good LB who works well within the Buffalo defense.  Is that the guy I want to pay top-5 LB money to when his contract expires?  Nope.  But he’s also not JAG.  The retention of Milano last offseason looks more and more like the steal of the offseason.  If they both stay healthy the Bills’ D will be just fine.

 

I agree with pretty much all of that.  
 

 

[BADOLBILZ]

55 minutes ago, GoBills808 said:

To that end (and I realize it's getting off-topic but still relevant vis a vis a player's skillset, real or imagined): this has been studied by several outlets, FootballOutsiders as referenced before as well as FootballPerspective and others. It may seem counterintuitive to you but there is no predictive data to be derived from fumble recoveries. It is entirely random and using them as proof of a player's relative abilities isn't going to offer anything in the way of insight. It's a totally useless endeavor.

 

If fumble recoveries were a skill and not luck, we would expect a nonrandom set of values of a linear regression between the relationship of a team's fumble recoveries from the first half of the season to the second. Instead, you get this:

 

...ie a totally random distribution. The relationship between a team's fumble recoveries from the first half of the season to the second is <.02, or a  nonexplanatory independent variable. Contrast this to the observed values of a team's yards per carry, which IS predictive over time, distributed over the same first half/second half of the season model:

 

 

...you get a very neat linear progression. tldr fumble recoveries are known quantities , entirely random, and should not be used to infer the quality (or lack thereof) of a player/defense/team

 

 

This is the same mistake The Big Cat made with his "heat maps" when defending Rex Ryan's defense in Buffalo............trying to explain away an inherent deficiency using a statistical field that does not take into account individual player strengths and weaknesses.

 

While team fumble recoveries.........and to some extent turnovers.........can seem random or fluctuate greatly from season to season.........certain players are clearly exceptional at creating them AND coming away with the football.    Undeniably.    

 

[GoBills808]

6 minutes ago, BADOLBILZ said:

 

 

This is the same mistake The Big Cat made with his "heat maps" when defending Rex Ryan's defense in Buffalo............trying to explain away an inherent deficiency using a statistical field that does not take into account individual player strengths and weaknesses.

 

While team fumble recoveries.........and to some extent turnovers.........can seem random or fluctuate greatly from season to season.........certain players are clearly exceptional at creating them AND coming away with the football.    Undeniably.    

 

No.

 

https://www.footballoutsiders.com/stat-analysis/2004/guest-column-turnovers-and-unpredictability-defense

[BADOLBILZ]

1 minute ago, GoBills808 said:

No.

 

https://www.footballoutsiders.com/stat-analysis/2004/guest-column-turnovers-and-unpredictability-defense

 

Yes.

 

https://www.pro-football-reference.com/players/T/TillCh20.htm

[GoBills808]

5 minutes ago, BADOLBILZ said:

 

Yes.

 

https://www.pro-football-reference.com/players/T/TillCh20.htm

So he must have the most forced fumbles of all time then…

[Dopey]

3 hours ago, Alphadawg7 said:

You clearly missed the sarcasm in using the same extremities that I was countering.  
 

And he executed that play perfectly, and NOT all LBs do that, Beasley makes LBs look silly on the same route regularly.  And he did this on more than just this play.  
 

Sorry, you’re still downplaying good plays for the purpose of exaggerating bad plays to confirm your opinion of him.

 

There isn’t a GM in football who wouldn’t have picked up Edmunds option just like Beane did.  
 

This is IMHO the worst thread of the year full of over exaggerated hot takes and non sense.  I don’t know if I’ve ever seen a thread with more blatant confirmation bias attempts as this one.  

Edmunds is our 2021 whipping boy.

[Simon]

Some guys are just ball magnets and that's the way it is. Like a great rebounder who just instinctually senses where to set up and box out, some guys have a natural tendency to take the ball away because they know where it is and sense where it's about to be.. Maybe it's because they're so anticipatory or they're seeing the game at a different speed than everybody else, whatever the case it's real and no mathematical rationalization is going to account for it. Watch Tre White or Milano making tackles; they've always got one fist punching in the process of wrapping up. Or Poyer's instinctive timing to be able to coil and then attack at the vulnerable moment.

 

3 hours ago, BADOLBILZ said:

I was reminded of the Poyer/Milano strip/recovery that was overturned at the very end of the Colts playoff game.

 

That's a perfect example. Poyer could have absolutely walloped Pascal as he was stupidly trying to get back up, but he didn't. He instinctively held himself up to avoid contact until Pascal was off the ground and then hammered the ball the split second that Pascal's knee came off the turf. You can't teach guys to see the game at that kind of a different speed than everybody else, you either have that or you don't. And Edmunds just doesn't. 

[YattaOkasan]

1 minute ago, Simon said:

Some guys are just ball magnets and that's the way it is. Like a great rebounder who just instinctually senses where to set up and box out, some guys have a natural tendency to take the ball away because they know where it is and sense where it's about to be.. Maybe it's because they're so anticipatory or they're seeing the game at a different speed than everybody else, whatever the case it's real and no mathematical rationalization is going to account for it. Watch Tre White or Milano making tackles; they've always got one fist punching in the process of wrapping up. Or Poyer's instinctive timing to be able to coil and then attack at the vulnerable moment.

 

 

That's a perfect example. Poyer could have absolutely walloped Pascal as he was stupidly trying to get back up, but he didn't. He instinctively held himself up to avoid contact until Pascal was off the ground and then hammered the ball the split second that Pascal's knee came off the turf. You can't teach guys to see the game at that kind of a different speed than everybody else, you either have that or you don't. And Edmunds just doesn't. 

Think they are talking about fumble recovery which is random by my understanding.  Don’t know that anyone has metrics on recovering fumbles relative to opportunities. But which team gets it is pretty much 50/50

[Simon]

Just now, YattaOkasan said:

Think they are talking about fumble recovery which is random by my understanding.  Don’t know that anyone has metrics on recovering fumbles relative to opportunities. But which team gets it is pretty much 50/50

It may be 50/50 by teams, but there are individuals who excel in that area because of how they see the game. Some guys just recognize before others when the ball is going to come out, or what angle at which it's going to hit the turf and bounce or when they can handle it with a scoop or need to get down on it because somebody else is closing behind them, etc.

[Coach Tuesday]

Lost in all of this Edmunds discussion is the fact that Milano through three games is playing at All Pro level.

[YattaOkasan]

15 minutes ago, Simon said:

It may be 50/50 by teams, but there are individuals who excel in that area because of how they see the game. Some guys just recognize before others when the ball is going to come out, or what angle at which it's going to hit the turf and bounce or when they can handle it with a scoop or need to get down on it because somebody else is closing behind them, etc.

I just don’t know that this can be proven. Is there anything to support. All the other data suggests it’s random. 

[Simon]

2 minutes ago, YattaOkasan said:

I just don’t know that this can be proven. Is there anything to support. All the other data suggests it’s random. 

It's not something you calculate, I think it's just something you see.

[YattaOkasan]

8 minutes ago, Simon said:

It's not something you calculate, I think it's just something you see.

Would like some data. However, I am arguing there are things about Edmunds we’re not seeing; though I am using facts to support.  I wait for PFF up to come up with a metric 😅

[Simon]

6 minutes ago, YattaOkasan said:

Would like some data. However, I am arguing there are things about Edmunds we’re not seeing; though I am using facts to support.  I wait for PFF up to come up with a metric 😅

Not everything can be quantified; at least in a manner which we're capable of understanding.

 

As for Edmunds, he's just sort of OK at everything. Aside from the occasional play where he looks sort of silly, he does an adequate job with most of his responsibilities. I don't think he's near as good an athlete as everybody wants to give him credit for, nor do I think he's a glaring hole in the Bills D.

I just don't want to see us lose a really good player (Poyer, Oliver, Dawkins, etc) because we decided to tie up a bunch of cap space on a guy who can be replaced by a cheaper player who is just.....OK.

[Southern_Bills]

2 hours ago, eball said:

 

I’ve come to the conclusion that Edmunds is a pretty good LB who works well within the Buffalo defense.  Is that the guy I want to pay top-5 LB money to when his contract expires?  Nope.  But he’s also not JAG.  The retention of Milano last offseason looks more and more like the steal of the offseason.  If they both stay healthy the Bills’ D will be just fine.

 

 

One of the most accurate statements about Edmunds. He doesn't have to be Ray Lewis.

[YattaOkasan]

5 minutes ago, Simon said:

Not everything can be quantified; at least in a manner which we're capable of understanding.

 

As for Edmunds, he's just sort of OK at everything. Aside from the occasional play where he looks sort of silly, he does an adequate job with most of his responsibilities. I don't think he's near as good an athlete as everybody wants to give him credit for, nor do I think he's a glaring hole in the Bills D.

I just don't want to see us lose a really good player (Poyer, Oliver, Dawkins, etc) because we decided to tie up a bunch of cap space on a guy who can be replaced by a cheaper player who is just.....OK.

I feel very confident beane has a path for keeping all these players (Dawkins and Poyer have been extended already) whether it includes edmunds or not. 

[BillsLux]

For Tremaine defenders,  I beg you to spend some time during the next game to really concentrate on watching him.  He rarely gets off a block, he never makes a stop on a runner coming at him, and the most potent plays opposing teams have against us is a crossing pattern in front of him.

 

The argument that our middle linebackers role to free up Milano is tough to take seriously. We have Star up the middle who eats up space so linebackers can make plays, and now I'm supposed to believe that we have a middle linebacker whose roll is to allow Milano to make plays.

 

The problem is that he might be tall and he might be fast in the 100 yard dash, but he look doesn't look at all quick or shifty.

[Coach Tuesday]

2 hours ago, GoBills808 said:

No.

 

https://www.footballoutsiders.com/stat-analysis/2004/guest-column-turnovers-and-unpredictability-defense

I’ve read that article before, but I went back and read it more closely tonight after you linked to it.

 

I’m no statistician but it strikes me as problematic.  If I understand correctly what he’s done, he’s measured teams over two-year periods to look for correlation between what he calls “drive rate” (which appears to be some kind of accumulation of points, yards and turnovers per drive) and measured each to test whether forced turnovers correlate to drive rate.  He concludes that the variance/unpredictability of defensive turnovers between years 1 and 2 demonstrates that turnovers are “random,” because “if a particular measure closely represents actual ability, then that stat will hold relatively consistent from one year to the next.”

 

The data set may be flawed - 2 years is both too short, in that there may not be enough turnovers to measure correlation, and too long, in that - as he notes - teams change out players every year.  
 

But also, I just don’t understand how he gets to his conclusion that a high variance from year 1 to year 2 means it’s random.  There could be a multitude of factors explaining the discrepancy; there could be some additional factor he’s not accounting for, etc.  He never fully explains it other than to say, “by far the least predictable of these drive rate stats is the defensive turnover per drive rate. This suggests that defenses have relatively little persistent ability to force turnovers.”  Does it?  Couldn’t it mean that defenses that fail to produce turnovers might excel at other things resulting in “successful” drive rates (however he’s measuring it), such as keeping the play in front of them and making tackles (Jauron Ball)?  Maybe forcing and recovering fumbles is really hard, and a great defense that fails to force a fumble on second down is able to get a drive-ending sack on third down instead.  We need more information and analysis than what he’s put forth in this article.
 

And finally, he concludes by remarking, “Whether there are some individual players with a special ability to force turnovers significantly above average rates would be an interesting subject for further study.”  Note that if this is a possibility, the converse is also possible: perhaps there are some individual players with a below-average ability to cause turnovers.

 

Which brings us full-circle…

Edited September 28, 2021 by Coach Tuesday

[Southern_Bills]

14 minutes ago, ScottLaw said:

You also aren’t paying him 13-15 million per year when in all likelihood the defense doesn’t miss a beat with his replacement making less than half that amount…. 

 

I'm not aware of anyone, and I mean anyone, advocating for top 3 LB money for Edmunds. 

 

Perhaps I've missed that conversation or it's just something are thinking up in there head, who knows.

[GoBills808]

15 minutes ago, Coach Tuesday said:

I’ve read that article before, but I went back and read it more closely tonight after you linked to it.

 

I’m no statistician but it strikes me as problematic.  If I understand correctly what he’s done, he’s measured teams over two-year periods to look for correlation between what he calls “drive rate” (which appears to be some kind of accumulation of points, yards and turnovers per drive) and measured each to test whether forced turnovers correlate to drive rate.  He concludes that the variance/unpredictability of defensive turnovers between years 1 and 2 demonstrates that turnovers are “random,” because “if a particular measure closely represents actual ability, then that stat will hold relatively consistent from one year to the next.”

 

The data set may be flawed - 2 years is both too short, in that there may not be enough turnovers to measure correlation, and too long, in that - as he notes - teams change out players every year.  
 

But also, I just don’t understand how he gets to his conclusion that a high variance from year 1 to year 2 means it’s random.  There could be a multitude of factors explaining the discrepancy; there could be some additional factor he’s not accounting for, etc.  He never fully explains it other than to say, “by far the least predictable of these drive rate stats is the defensive turnover per drive rate. This suggests that defenses have relatively little persistent ability to force turnovers.”  Does it?  Couldn’t it mean that defenses that fail to produce turnovers might excel at other things resulting in “successful” drive rates (however he’s measuring it), such as keeping the play in front of them and making tackles (Jauron Ball)?  Maybe forcing and recovering fumbles is really hard, and a great defense that fails to force a fumble on second down is able to get a drive-ending sack on third down instead.  We need more information and analysis than what he’s put forth in this article.
 

And finally, he concludes by remarking, “Whether there are some individual players with a special ability to force turnovers significantly above average rates would be an interesting subject for further study.”  Note that if this is a possibility, the converse is also possible: perhaps there are some individual players with a below-average ability to cause turnovers.

 

Which brings us full-circle…

 

22 minutes ago, Coach Tuesday said:

I’ve read that article before, but I went back and read it more closely tonight after you linked to it.

 

I’m no statistician but it strikes me as problematic.  If I understand correctly what he’s done, he’s measured teams over two-year periods to look for correlation between what he calls “drive rate” (which appears to be some kind of accumulation of points, yards and turnovers per drive) and measured each to test whether forced turnovers correlate to drive rate.  He concludes that the variance/unpredictability of defensive turnovers between years 1 and 2 demonstrates that turnovers are “random,” because “if a particular measure closely represents actual ability, then that stat will hold relatively consistent from one year to the next.”

 

The data set may be flawed - 2 years is both too short, in that there may not be enough turnovers to measure correlation, and too long, in that - as he notes - teams change out players every year.  
 

But also, I just don’t understand how he gets to his conclusion that a high variance from year 1 to year 2 means it’s random.  There could be a multitude of factors explaining the discrepancy; there could be some additional factor he’s not accounting for, etc.  He never fully explains it other than to say, “by far the least predictable of these drive rate stats is the defensive turnover per drive rate. This suggests that defenses have relatively little persistent ability to force turnovers.”  Does it?  Couldn’t it mean that defenses that fail to produce turnovers might excel at other things resulting in “successful” drive rates (however he’s measuring it), such as keeping the play in front of them and making tackles (Jauron Ball)?  Maybe forcing and recovering fumbles is really hard, and a great defense that fails to force a fumble on second down is able to get a drive-ending sack on third down instead.  We need more information and analysis than what he’s put forth in this article.
 

And finally, he concludes by remarking, “Whether there are some individual players with a special ability to force turnovers significantly above average rates would be an interesting subject for further study.”  Note that if this is a possibility, the converse is also possible: perhaps there are some individual players with a below-average ability to cause turnovers.

 

Which brings us full-circle…

It's not two years, its two year increments over six years. That's 155 pairs of consecutive team-seasons, or 310 individual data points.

[BADOLBILZ]

2 hours ago, GoBills808 said:

So he must have the most forced fumbles of all time then…

 

Tillman has forced the most fumbles by a back 7 player since it became a stat, which is what we are talking about, yes.

 

The technique that White used to cause that fumble Sunday is named after him for a reason.

 

[BillsFan130]

Edmunds is definitely a polarizing player to say the least lol, 23 pages and counting.

 

For me, I can't say I'm a big fan of him.

 

Thought he was really good in 2019.

 

Thought he struggled pretty badly last year. (although injuries played a role)

 

And this year I think he played well week 1, and didn't play very well the past 2 games.

 

In saying all of this, I can't blame Beane picking up his 5th year option. He has shown he can be the guy, I just don't think he's done it consistently enough. 

 

 

Edited September 28, 2021 by BillsFan130

[GoBills808]

12 minutes ago, BADOLBILZ said:

 

Tillman has forced the most fumbles by a back 7 player since it became a stat, which is what we are talking about, yes.

 

The technique that White used to cause that fumble Sunday is named after him for a reason.

 

Actually we were talking about fumble RECOVERIES, but since you insist: guess who has more forced fumbles on a per game basis, Tremaine Edmunds or Matt Milano?   😂😂😂

 

Look if you guys want to believe in things that don't exist, go ahead. There is no data to suggest that fumble recoveries are anything other than random.

 

[Thurman#1]

9 hours ago, Coach Tuesday said:

He did not “blanket” anything.  They were coached yesterday to take away Heineken Light’s first read.  Edmunds on that play is shading to the inside bracket of the first read.  He literally takes like three steps to his left.  Tre White clone?  It’s hardly an example of a game-changing play, it’s just good well-coached team defense.  

Meanwhile Milano recovered another fumble yesterday…

 

 

He absolutely did blanket that route. What he did is the definition of blanketing. Blanketing is simply covering something very well. That's what Edmunds did. The fact that you're denying it says more about how desperate you are to find negatives about Edmunds than it does about his play. Heinicke wanted to go there. The reason he didn't is that the route had been very effectively covered. Blanketed.

 

When they cut back at the end as the ball goes towards the INT you can see that Edmunds is still about a yard away from Humphries, even as his focus flows towards the side of the field the ball is headed towards.

 

As for the rest of it, you are guessing. What we know is that he covered a zone, but that he very quickly got to just the right spot. This likely came about through a ton of film study, through listening to good coaching, to his instincts in coverage and to an excellent grasp of what Washington was doing on the play and an understanding of what his part was in thwarting them. Not to mention being 6'5" with an 83-inch wingspan, a silhouette that makes QBs sweat at the idea of throwing near.

 

He's a really good coverage defender, and this was an example of him doing everything right, and of good team defense, and of Edmunds doing his job beautifully, which forced a longer time in the pocket, which allowed the Bills to make a big play.

[BADOLBILZ]

12 minutes ago, GoBills808 said:

Actually we were talking about fumble RECOVERIES, but since you insist: guess who has more forced fumbles on a per game basis, Tremaine Edmunds or Matt Milano?   😂😂😂

 

Look if you guys want to believe in things that don't exist, go ahead. There is no data to suggest that fumble recoveries are anything other than random.

 

 

Yuck it up Mr. "nobody understands statistics like me".

 

Tremaine Edmunds has played about 400 more defensive snaps in his career than Matt Milano. 😙

 

That's about the equivalent of 6 more full games of snaps.

 

Not remembering that Milano has missed A LOT of time over his career shows a real lack of awareness on your part..........uninstinctive posting.

 

Milano has come away with 12 turnovers in less snaps than Edmunds has taken to produce just 3.  

 

Keep thinking that all turnovers are random though...........turnover differential is arguably the most decisive team stat in the NFL so I guess winning is random as well. 

 

[Thurman#1]

1 hour ago, BillsLux said:

For Tremaine defenders,  I beg you to spend some time during the next game to really concentrate on watching him.  He rarely gets off a block, he never makes a stop on a runner coming at him, and the most potent plays opposing teams have against us is a crossing pattern in front of him.

 

The argument that our middle linebackers role to free up Milano is tough to take seriously. We have Star up the middle who eats up space so linebackers can make plays, and now I'm supposed to believe that we have a middle linebacker whose roll is to allow Milano to make plays.

 

The problem is that he might be tall and he might be fast in the 100 yard dash, but he look doesn't look at all quick or shifty.

 

 

We have watched. What you have there is not particularly correct. He often makes stops on runners. Do the runners generally try to run away from Tremaine? Yes, they do, it's their nature to run to holes if they can, but he fills holes and makes tackles plenty.

 

Fair enough that he doesn't look quick or shifty. Tall guys with long legs rarely do. But the facts show that however he looks, he actually is extremely athletic. His scores show that, as NewEra displayed above. I would agree with you this far, when he has to reverse field, he looks a bit awkward, but he put himself in the right place often enough that you rarely see him reversing field.

[Thurman#1]

27 minutes ago, BADOLBILZ said:

 

Yuck it up Mr. "nobody understands statistics like me".

 

Tremaine Edmunds has played about 400 more defensive snaps in his career than Matt Milano. 😙

 

That's about the equivalent of 6 more full games of snaps.

 

Not remembering that Milano has missed A LOT of time over his career shows a real lack of awareness on your part..........uninstinctive posting.

 

Milano has come away with 12 turnovers in less snaps than Edmunds has taken to produce just 3.  

 

Keep thinking that all turnovers are random though...........turnover differential is arguably the most decisive team stat in the NFL so I guess winning is random as well. 

 

 

 

Typical. When there's no reasonable argument to be made, pull a quick switch, insert a straw man and pretend you're arguing with what he actually said.

 

Remind us, with your arguments about turnover differential, did he ever say that turnovers are random? Yeah, the answer to that is "No." Wouldn't make sense to. Interceptions are anything but random. Nor is causing fumbles. What is pretty close to random is who recovers them. 

 

Having disposed of your first straw man, let's proceed to the second. Did he say that recovering fumbles was evenly distributed among all players based on number of snaps? Because if he had, you certainly polished off that argument very well. Thing is, that's not what he said.

 

I wouldn't go quite so far as random myself, not on the player vs. player level. Some positions get more, some get less, those are facts. On the team level, yeah, the stats show that offense/defense recoveries go pretty close to 50:50. On the player vs. player level, though, yeah, it's wildly affected by luck, by how close you are, by direction you're going, by the bounce, and so on. That will - duh - not result in recoveries being spread out evenly across the roster based on number of snaps. That's not how randomness works.

Edited September 28, 2021 by Thurman#1

[Norcalbillsfan]

On 9/26/2021 at 9:14 PM, NewEra said:

Regarding Edmunds, I agree 💯

 

I think the premise of people “turning on Josh” is a bit overstated.  He had one of the best years in league history last season.  Going from what we saw last season to what we saw week 1 and 2  caught the entire fan base (and media) off guard.  He IS our franchise.  We talked about him not playing well but I can’t recall anyone turning on 

"If Allen Plays Like Crap This Week, Accept that He's Simply Not Good

Wednesday at 07:43 AM 

 

It was literally a thread, and quite a few people agreed lol. But I do admit many people were just worried, most weren't grabbing pitch forks.

[Norcalbillsfan]

23 hours ago, HappyDays said:

 

Having just rewatched the game a couple hours ago, Edmunds was seemingly responsible for a large chunk of Washington's passing yardage. The secondary did their job, the d-line did their job, Milano did his job. When Washington moved the ball it was typically because Edmunds was covering grass or moving a split second too late.

 

 

I'm saying his weaknesses are so easily exploitable that teams who know how to attack him can erase his positive attributes and move the ball at will. He is lucky to be playing in front of an elite secondary and behind a much improved front four. That makes it easier to hide his flaws.

Thank you, I thought I was the only one that noticed him covering grass on large chunk plays. But to me the biggest frustration is as a MLB getting absolutely put on roller skates on screen plays by an open feild block by slow lineman or tight ends. 

And many seem to forget how exploitable he his when Milano wasn't on the feild last year, the LB core without Milano got diced up in short passing.

[NewEra]

19 minutes ago, Norcalbillsfan said:

"If Allen Plays Like Crap This Week, Accept that He's Simply Not Good

Wednesday at 07:43 AM 

 

It was literally a thread, and quite a few people agreed lol. But I do admit many people were just worried, most weren't grabbing pitch forks.

Yeah.  It was a thread started by filthybeast.  I didn’t think that counted. One person “agreed” with the OP.  Guess who?  The one and only.  I don’t think anyone turned.  Fans are emotional filthy beasts.