Today's Analytics Presser... Sashi And Paul Take The Stage
-
regarding Manziel we don't need a reason to dump him. Regarding a new GM it sounds like the titles will be VP of Player Personell. Can the case be made that's a step up from GM?After the MKC question about Johnny, sounds like we cant dump him because hes a drunk...just like you cant fire a guy because he is getting help. Easiest 8 mil of your life. If he was smart and invested it, he would never have to work again.
regarding Manziel we don't need a reason to dump him. Regarding a new GM it sounds like the titles will be VP of Player Personell. Can the case be made that's a step up from GM?
It sounds that way to me as well. With the timing & quietness of this search, I do wonder about Mr Manning. Just a thought.
Mike
-
Idk how I feel about the power structure. Sachi Brown has final say? Idk...
Idk how I feel about the power structure. Sachi Brown has final say? Idk...
Hard to know at this point isn't it? There is something to be said for an impartial arbiter weighing the arguments should a consensus be unreachable, but I don't think it'll come to that. I think the analytics will break "ties".
Mark Dominic gave a great example of an analytic yesterday on NFL Insiders. In evaluating QB prospects accuracy the most used stat is completion percentage adjusted for drops. An analytic measure of accuracy would score each throw rating it from:
- 1 for a dead-on, in-stride, chest-high throw, to;
- 5 for a completely out-of-reach throw be it high, wide or short.
The gradation between the extremes would be critical and certainly could vary from system to system, but you could see:
- 2 for a pass that causes a slight break in stride and/or a small leap or bend and/or a reach; and
- 4 for a pass requiring a stop and leap, or a dive.
That would leave only 3 to be defined. Maybe a reach behind? At the limit of a player's catch radius?
Special allowances might be made for passes thrown intentionally high or low, e.g., in goal-line or quick slant situations. Similarly a back-shoulder pass could be rewarded. Flagging such throws for group review would be appropriate to eliminate bias. The impact of getting some wrong, either way, will be reduced as the number of graded throws rises.
Note that the placement of the pass, the accuracy with which it is delivered is what is measured by the above. Whether a given pass is ultimately completed, dropped or broken up is immaterial. It's this last bit that really got my attention. Why should a QB be rewarded at all for a completion where his very tall WR hauls down a jump ball? Conversely why should that QB be penalized for a WR's failure to separate coming out of a break he was unable to sell and the DB gets a hand in to prevent a completion?
It's clear to me how much more revealing the accuracy measure produced by the above would be than drop-adjusted completion percentages.
Anyway then passes can be graded and averaged overall or by areas of the field with 1.0 being the ideal. Teams can then emphasize the areas most important to them in the Offense they run in choosing among QB options.
It seems like a great tie-breaker tool...
Of course it's not always going to be a matter of two QBs tying. So ideally some kind of "overall value to the team" metric has to evolve. Perhaps it is not unlike the player grades found at NFL.com. It may well even roll that number into the calculation. But I doubt that system exists today.
"When you have a shared vision it is a lot easier to make unpopular decisions." - Paul DePodesta "When you win, say nothing. When you lose, say less." - Paul Brown Complete 2017 Browns Board Mock: http://thebrownsboard.com/forums/index.php?showtopic=46196 -
Idk how I feel about the power structure. Sachi Brown has final say? Idk...
Hard to know at this point isn't it? There is something to be said for an impartial arbiter weighing the arguments should a consensus be unreachable, but I don't think it'll come to that. I think the analytics will break "ties".
Mark Dominic gave a great example of an analytic yesterday on NFL Insiders. In evaluating QB prospects accuracy the most used stat is completion percentage adjusted for drops. An analytic measure of accuracy would score each throw rating it from:
- 1 for a dead-on, in-stride, chest-high throw, to;
- 5 for a completely out-of-reach throw be it high, wide or short.
The gradation between the extremes would be critical and certainly could vary from system to system, but you could see:
- 2 for a pass that causes a slight break in stride and/or a small leap or bend and/or a reach; and
- 4 for a pass requiring a stop and leap, or a dive.
That would leave only 3 to be defined. Maybe a reach behind? At the limit of a player's catch radius?
Special allowances might be made for passes thrown intentionally high or low, e.g., in goal-line or quick slant situations. Similarly a back-shoulder pass could be rewarded. Flagging such throws for group review would be appropriate to eliminate bias. The impact of getting some wrong, either way, will be reduced as the number of graded throws rises.
Note that the placement of the pass, the accuracy with which it is delivered is what is measured by the above. Whether a given pass is ultimately completed, dropped or broken up is immaterial. It's this last bit that really got my attention. Why should a QB be rewarded at all for a completion where his very tall WR hauls down a jump ball? Conversely why should that QB be penalized for a WR's failure to separate coming out of a break he was unable to sell and the DB gets a hand in to prevent a completion?
It's clear to me how much more revealing the accuracy measure produced by the above would be than drop-adjusted completion percentages.
Anyway then passes can be graded and averaged overall or by areas of the field with 1.0 being the ideal. Teams can then emphasize the areas most important to them in the Offense they run in choosing among QB options.
It seems like a great tie-breaker tool...
Of course it's not always going to be a matter of two QBs tying. So ideally some kind of "overall value to the team" metric has to evolve. Perhaps it is not unlike the player grades found at NFL.com. It may well even roll that number into the calculation. But I doubt that system exists today.
-
lol... unfortunately the rest of my post was about how bad that stat is...
"When you have a shared vision it is a lot easier to make unpopular decisions." - Paul DePodesta "When you win, say nothing. When you lose, say less." - Paul Brown Complete 2017 Browns Board Mock: http://thebrownsboard.com/forums/index.php?showtopic=46196 -
lol... unfortunately the rest of my post was about how bad that stat is...
I think it's accurate in that particular case
But maybe not as much for Lynch - who is 75% adjusted for drops. Given the fact he only averaged 4 20+ yard passes per game (I think) and a few other things. The simple stat probably doesn't paint the best picture on accuracy.
I think your scale is better, maybe something more advanced like that is part of the service the Browns paid $100,000 for to ignore.
-
DePo has an assignment that I think we saw coming...
Cleveland Browns to analyze NFL's best turnarounds
Mathew Florjancic 5:00 AM. EST February 02, 2016
CLEVELAND -- With the hiring of former Major League Baseball executive Paul DePodesta as Chief Strategy Officer, who is known for his knowledge and application of analytics when it comes to making decisions about the franchise, the Cleveland Browns are changing the course of action for a struggling organization that has been out of the postseason since 2002.
In addition to analyzing all of the players on the roster, the Browns are going to evaluate how teams around the NFL have turned around their fortunes in short order.
http://www.wkyc.com/sports/nfl/browns/cleveland-browns-to-analyze-nfls-best-turnarounds/28686513
-
What if the conclusion is, "They never hired a baseball guy to analyze football"?
-
What if the conclusion is, "They never hired a baseball guy to analyze football"?
What if the conclusion is, "They never hired a baseball guy to analyze football"?
lol... of course, that also applies to teams that have never turned around...
-
-
History shall be made this year folks
I don't know, but it should be fun to watch how it all plays out. I suspect that we will see picks that stick to most boards. I doubt we will see guys thought of as 6th rounders being picked in round 3 or anything like that. I think the process will shine by not allowing one guys ego to interfere with making sound football decisions.
Based on what we have seen from "football" guys the last 10 years, why not see if crunching numbers helps us find a edge? I don't think one needs to be a football guy to identify good football players. Us as fans can usually spot guys who are better than other players, and we aren't football guys.
The scouts will file reports. They are football guys. They will put a grade on hundreds of players. Then cross checkers will see if they agree or not on any who's grade may not be in line with conventional wisdom. Teams also employ independant scouting services that all the teams can tap in to...a form of cross checking. There are plenty of "football" eyes on all of these players before our front office starts making decisions. It doesn't take a football guy to make those decisions. It take a guy or team of guys to make a logical decision and keep personal desire to rule the day. This will prevent moments like Mike Holmgren stepping in, exerting his desire and grabbing Brandon Weeden when we did when all the information we amassed indicated we could wait to make that pick.
Holmgren allowed his desire to fill a QB spot intefere with making a sound football decision.
If everybody had like minds, we would never learn. -
History shall be made this year folks
I don't know, but it should be fun to watch how it all plays out. I suspect that we will see picks that stick to most boards. I doubt we will see guys thought of as 6th rounders being picked in round 3 or anything like that. I think the process will shine by not allowing one guys ego to interfere with making sound football decisions.
Based on what we have seen from "football" guys the last 10 years, why not see if crunching numbers helps us find a edge? I don't think one needs to be a football guy to identify good football players. Us as fans can usually spot guys who are better than other players, and we aren't football guys.
The scouts will file reports. They are football guys. They will put a grade on hundreds of players. Then cross checkers will see if they agree or not on any who's grade may not be in line with conventional wisdom. Teams also employ independant scouting services that all the teams can tap in to...a form of cross checking. There are plenty of "football" eyes on all of these players before our front office starts making decisions. It doesn't take a football guy to make those decisions. It take a guy or team of guys to make a logical decision and keep personal desire to rule the day. This will prevent moments like Mike Holmgren stepping in, exerting his desire and grabbing Brandon Weeden when we did when all the information we amassed indicated we could wait to make that pick.
Holmgren allowed his desire to fill a QB spot intefere with making a sound football decision.
I don't know, but it should be fun to watch how it all plays out. I suspect that we will see picks that stick to most boards. I doubt we will see guys thought of as 6th rounders being picked in round 3 or anything like that. I think the process will shine by not allowing one guys ego to interfere with making sound football decisions.
Based on what we have seen from "football" guys the last 10 years, why not see if crunching numbers helps us find a edge? I don't think one needs to be a football guy to identify good football players. Us as fans can usually spot guys who are better than other players, and we aren't football guys.
The scouts will file reports. They are football guys. They will put a grade on hundreds of players. Then cross checkers will see if they agree or not on any who's grade may not be in line with conventional wisdom. Teams also employ independant scouting services that all the teams can tap in to...a form of cross checking. There are plenty of "football" eyes on all of these players before our front office starts making decisions. It doesn't take a football guy to make those decisions. It take a guy or team of guys to make a logical decision and keep personal desire to rule the day. This will prevent moments like Mike Holmgren stepping in, exerting his desire and grabbing Brandon Weeden when we did when all the information we amassed indicated we could wait to make that pick.
Holmgren allowed his desire to fill a QB spot intefere with making a sound football decision.
I completely agree. They're going to be making decisions based on a wide variety of factors and there won't be someone's ego making the choice and that should avoid the Weeden and Manziel and other mistakes that have been made over the years.
-
I don't know, but it should be fun to watch how it all plays out. I suspect that we will see picks that stick to most boards. I doubt we will see guys thought of as 6th rounders being picked in round 3 or anything like that. I think the process will shine by not allowing one guys ego to interfere with making sound football decisions.
Based on what we have seen from "football" guys the last 10 years, why not see if crunching numbers helps us find a edge? I don't think one needs to be a football guy to identify good football players. Us as fans can usually spot guys who are better than other players, and we aren't football guys.
The scouts will file reports. They are football guys. They will put a grade on hundreds of players. Then cross checkers will see if they agree or not on any who's grade may not be in line with conventional wisdom. Teams also employ independant scouting services that all the teams can tap in to...a form of cross checking. There are plenty of "football" eyes on all of these players before our front office starts making decisions. It doesn't take a football guy to make those decisions. It take a guy or team of guys to make a logical decision and keep personal desire to rule the day. This will prevent moments like Mike Holmgren stepping in, exerting his desire and grabbing Brandon Weeden when we did when all the information we amassed indicated we could wait to make that pick.
Holmgren allowed his desire to fill a QB spot intefere with making a sound football decision.
I completely agree. They're going to be making decisions based on a wide variety of factors and there won't be someone's ego making the choice and that should avoid the Weeden and Manziel and other mistakes that have been made over the years.
I completely agree. They're going to be making decisions based on a wide variety of factors and there won't be someone's ego making the choice and that should avoid the Weeden and Manziel and other mistakes that have been made over the years.
And of course the ego that has to be suppressed most of all is that of the owner. But then he is the guy that brought all these guys in....so, maybe: Mission Accomplished?
(Premature Ejaculatory statement? Just like W?)
-
History shall be made this year folks
I don't know, but it should be fun to watch how it all plays out. I suspect that we will see picks that stick to most boards. I doubt we will see guys thought of as 6th rounders being picked in round 3 or anything like that. I think the process will shine by not allowing one guys ego to interfere with making sound football decisions.
Based on what we have seen from "football" guys the last 10 years, why not see if crunching numbers helps us find a edge? I don't think one needs to be a football guy to identify good football players. Us as fans can usually spot guys who are better than other players, and we aren't football guys.
The scouts will file reports. They are football guys. They will put a grade on hundreds of players. Then cross checkers will see if they agree or not on any who's grade may not be in line with conventional wisdom. Teams also employ independant scouting services that all the teams can tap in to...a form of cross checking. There are plenty of "football" eyes on all of these players before our front office starts making decisions. It doesn't take a football guy to make those decisions. It take a guy or team of guys to make a logical decision and keep personal desire to rule the day. This will prevent moments like Mike Holmgren stepping in, exerting his desire and grabbing Brandon Weeden when we did when all the information we amassed indicated we could wait to make that pick.
Holmgren allowed his desire to fill a QB spot intefere with making a sound football decision.
Holmgren allowed his desire to fill a QB spot intefere with making a sound football decision.
And how does that apply now as we see the lean of the Browns towards going with a QB @ #2 overall this early in the process and well before the analytics are done? They have not yet studied the new assignment too determine what made for turnarounds, but yet there has been a lot of info out there towards a QB. Hmmmmm. Will they really let the process make that decision or will they do another Holgren gut decision?
And will they look at key draft picks before the turnarounds? Hope so. I think they will find a lot of defensive picks there. But just my gut reaction.
-
Holmgren allowed his desire to fill a QB spot intefere with making a sound football decision.
And how does that apply now as we see the lean of the Browns towards going with a QB @ #2 overall this early in the process and well before the analytics are done? They have not yet studied the new assignment too determine what made for turnarounds, but yet there has been a lot of info out there towards a QB. Hmmmmm. Will they really let the process make that decision or will they do another Holgren gut decision?
And will they look at key draft picks before the turnarounds? Hope so. I think they will find a lot of defensive picks there. But just my gut reaction.
And how does that apply now as we see the lean of the Browns towards going with a QB @ #2 overall this early in the process and well before the analytics are done? They have not yet studied the new assignment too determine what made for turnarounds, but yet there has been a lot of info out there towards a QB. Hmmmmm. Will they really let the process make that decision or will they do another Holgren gut decision?
And will they look at key draft picks before the turnarounds? Hope so. I think they will find a lot of defensive picks there. But just my gut reaction.
I think it's a symptom of you either have a franchise QB or you're looking for one. If they think Goff (or Lynch or Wentz) is that guy, you have to take them at #2 this year. You have a guy in McCown who could hold the job down for the year if need be but is also very willing (his words and actions show this) to be the mentor and even the #2 guy to help a young QB along. I'd probably lean towards just throwing whoever it is out there week 1 and we can sink or swim with him this year.
-
Holmgren allowed his desire to fill a QB spot intefere with making a sound football decision.
And how does that apply now as we see the lean of the Browns towards going with a QB @ #2 overall this early in the process and well before the analytics are done? They have not yet studied the new assignment too determine what made for turnarounds, but yet there has been a lot of info out there towards a QB. Hmmmmm. Will they really let the process make that decision or will they do another Holgren gut decision?
And will they look at key draft picks before the turnarounds? Hope so. I think they will find a lot of defensive picks there. But just my gut reaction.
And how does that apply now as we see the lean of the Browns towards going with a QB @ #2 overall this early in the process and well before the analytics are done? They have not yet studied the new assignment too determine what made for turnarounds, but yet there has been a lot of info out there towards a QB. Hmmmmm. Will they really let the process make that decision or will they do another Holgren gut decision?
And will they look at key draft picks before the turnarounds? Hope so. I think they will find a lot of defensive picks there. But just my gut reaction.
Still a lot of time left to flesh out the study, but it is possible that step one was:
- Clear out all distractions.
-
Holmgren allowed his desire to fill a QB spot intefere with making a sound football decision.
And how does that apply now as we see the lean of the Browns towards going with a QB @ #2 overall this early in the process and well before the analytics are done? They have not yet studied the new assignment too determine what made for turnarounds, but yet there has been a lot of info out there towards a QB. Hmmmmm. Will they really let the process make that decision or will they do another Holgren gut decision?
And will they look at key draft picks before the turnarounds? Hope so. I think they will find a lot of defensive picks there. But just my gut reaction.
Will they really let the process make that decision or will they do another Holgren gut decision?
I share one thing with DePodesta. We both have a background in bioinformatics analysis. I can tell you in no uncertain terms that gut reactions are frowned upon at least explicitly trained out of anyone who is going to be performing any kind of statistical or informatics-based analysis.
-
Well these things are reassuring. So which teams do any of you think qualify for the study of those that have made relatively quick turnarounds?
-
2012 to 2013 Chiefs
2-14 to 11-5
New Coach - Reid
Trade for QB - Smith
Draft - OT, TE (Kelce), backup RB
FAs - FS, OT (re-sign), WR (re-sign). Vet WR, 2 LBs, and 2 CBs (and many other mid-level depth players)
-
2011 to 2012 Colts
2-14 to 11-5
New coach- Pagano
Draft #1 QB, 2 TEs, and a star WR
FAs - vet WR, NT, DE, and #1 CB
(77% I believe, impressive given the # of attempts and highest in class)
Hello! It looks like you're interested in this conversation, but you don't have an account yet.
Getting fed up of having to scroll through the same posts each visit? When you register for an account, you'll always come back to exactly where you were before, and choose to be notified of new replies (either via email, or push notification). You'll also be able to save bookmarks and upvote posts to show your appreciation to other community members.
With your input, this post could be even better π
Register Login