Revisiting the Khalil Mack Trade

[Sasquatch]

12 minutes ago, Pool said:

Attack the poster because you can't attack the argument. Classic. I can give you a lot of reasons why I think the Bears will be good next year. All you and @Sasquatch can come up with for them being bad is injuries. Laughable.

BS, I also pointed out that by losing your DC, it’s reasonable to expect an adjustment that could result in diminished performance.  Why you so insecure?  I realize it’s been a tough week to be a Bears fan, but you seem particularly edgy, LOL.

[MacReady]

6 minutes ago, Pool said:

Where are you getting 14m? This site says 19.7m. And that's before we cut Sims.

That is cap space, not effective cap space.  Effective cap space is the cap space you have AFTER you get to the minimum amount of players you actually have to have on a team.  The Bears have 19 million in cap space, but not effective cap space.  

You don't got 19 million.  Sorry.  I've already outlined that.  You're not keeping Amos and Callahan, lol.

https://overthecap.com/salary-cap-space/

Edited January 10, 2019 by Outpost31

[Pool]

1 minute ago, Sasquatch said:

BS, I also pointed out that by losing your DC, it’s reasonable to expect an adjustment that could result in diminished performance.  Why you so insecure?  I realize it’s been a tough week to be a Bears fan, but you seem particularly edgy, LOL.

We lost a DC on a very talented defense. I don't think any DC will have a whole lot of trouble getting production out of this group. And I'm not insecure. Just slightly annoyed. There's being a fan and then there's being a total homer (not directed at you btw).

[Pool]

1 minute ago, Outpost31 said:

That is cap space, not effective cap space.  Effective cap space is the cap space you have AFTER you get to the minimum amount of players you actually have to have on a team.  The Bears have 19 million in cap space, but not effective cap space.  

You don't got 19 million.  Sorry.  I've already outlined that.  You're not keeping Amos and Callahan, lol.

I never said we were keeping them both. I've actually stated a few times in this thread that we will lose one of them. Or are you just backtracking because you realize you are wrong and that we will be able to keep one?

[Sasquatch]

7 minutes ago, Pool said:

We lost a DC on a very talented defense. I don't think any DC will have a whole lot of trouble getting production out of this group. And I'm not insecure. Just slightly annoyed. There's being a fan and then there's being a total homer (not directed at you btw).

I get it.  I guess what we’re trying to say nicely (we’ll, sorta nicely, lol) is that there is data to support regression using predictive modeling and math.  Of course there’s no certainty - but literally every sector in the world benefits from some application of predictive, odds-based modeling.  You mentioned you’re not a math guy - cool - many folks aren’t- which makes this that much more important.

If you’re interested, check out these elementary/abridged descriptions of some useful mathematical models that are used in everyday life.  Hotels and airline industries use them to price-set based on optimal occupancy.  The medical industry uses them to predict the probability of a wide variety of medical conditions.  Scientists of course use them. War generals use them to predict their enemies options and moves.  Computer programs run algorithms with them.  It’s everywhere my friend.  And yes, they can even predict the likelihood of how many players in any given year will spend time on the injury list.

https://en.wikipedia.org/wiki/Logistic_regression

https://en.wikipedia.org/wiki/Monte_Carlo_method

For what it’s worth, I think the Bears will be formidable again next year.  But I do “predict” there will be some pains along the way.  So many factors to consider - tougher schedule, Packers and Vikings still formidable, new defensive coordinator, loss of players this off-season, possible injuries next season, and on and on.  Still, you have reason to be happy and to hope.  Not gonna take that away from you.

[Pool]

24 minutes ago, Sasquatch said:

I get it.  I guess what we’re trying to say nicely (we’ll, sorta nicely, lol) is that there is data to support regression using predictive modeling and math.  Of course there’s no certainty - but literally every sector in the world benefits from some application of predictive, odds-based modeling.  You mentioned you’re not a math guy - cool - many folks aren’t- which makes this that much more important.

If you’re interested, check out these elementary/abridged descriptions of some useful mathematical models that are used in everyday life.  Hotels and airline industries use them to price-set based on optimal occupancy.  The medical industry uses them to predict the probability of a wide variety of medical conditions.  Scientists of course use them. War generals use them to predict their enemies options and moves.  Computer programs run algorithms with them.  It’s everywhere my friend.  And yes, they can even predict the likelihood of how many players in any given year will spend time on the injury list.

https://en.wikipedia.org/wiki/Logistic_regression

https://en.wikipedia.org/wiki/Monte_Carlo_method

For what it’s worth, I think the Bears will be formidable again next year.  But I do “predict” there will be some pains along the way.  So many factors to consider - tougher schedule, Packers and Vikings still formidable, new defensive coordinator, loss of players this off-season, possible injuries next season, and on and on.  Still, you have reason to be happy and to hope.  Not gonna take that away from you.

Thanks. I get it. I think. I think the point I'm failing to get across was that the Bears could regress to the mean with their backups being injured and it wouldn't affect their overall record as much as losing a Hicks, Mack, or Trubisky. That it's not so much about predicting injuries but who gets injured. Does that make sense? And I appreciate you acknowledging that the Bears will still be good next year. I don't think they will get 12 wins again but I think 10 is pretty attainable and that they'll be fighting for the division again too. And there's no way GB will be as bad next year as they were this year. It was like the whole team imploded or something. And after all these years of hating GB I honestly just felt bad for your fans because I know how much it sucks to watch your team just fall apart. Besides, the rivalry is a lot more fun when both teams are good. It just doesn't happen that often lol

[Sasquatch]

3 minutes ago, Pool said:

Thanks. I get it. I think. I think the point I'm failing to get across was that the Bears could regress to the mean with their backups being injured and it wouldn't affect their overall record as much as losing a Hicks, Mack, or Trubisky. That it's not so much about predicting injuries but who gets injured. Does that make sense? And I appreciate you acknowledging that the Bears will still be good next year. I don't think they will get 12 wins again but I think 10 is pretty attainable and that they'll be fighting for the division again too. And there's no way GB will be as bad next year as they were this year. It was like the whole team imploded or something. And after all these years of hating GB I honestly just felt bad for your fans because I know how much it sucks to watch your team just fall apart. Besides, the rivalry is a lot more fun when both teams are good. It just doesn't happen that often lol

I actually followed your logic all along, but what I failed to articulate was that even if it’s just your JAGS, role-players, or backups that go down, it has an impact on the stars.  The stars need their supporting cast to be great.  Injuries test a team’s depth, which multiple draft picks - especially high draft picks - give you.  It’s all related.

Have a great off-season, and let’s hope for great football in the NFC North in 2019.

[vegas492]

10 hours ago, Pool said:

Where are you getting 14m? This site says 19.7m. And that's before we cut Sims.

Personally?  I think that there will be a good market for Callahan...more than $7M/year.  Chicago doesn't have the cap room to get him back if a market really develops.

Amos?  Safety.  That's been depressed as far as markets go.  I can see retaining him.  Maybe Massie too.  I don't see getting those 3 back

It is fun watching the NFL in general...and the Bears have a somewhat interesting blueprint going on with a rookie QB, rookie HC and a stud defender taking up a ton of cap space...with no picks.  If it works?  Who knows, maybe more teams will go down that road.  There is risk, for sure.  And there is surely reward, too.

I think if the Bears get back to 12 wins?  Well...that should be a tougher 12 wins than it was this year.  Getting to 12 again would be an amazing accomplishment.  

Here's one for yah, Pool.....To get to 12 wins, the magic number, you need to win them all at home and split on the road.  You see that happening?

[Pool]

36 minutes ago, vegas492 said:

Personally?  I think that there will be a good market for Callahan...more than $7M/year.  Chicago doesn't have the cap room to get him back if a market really develops.

Amos?  Safety.  That's been depressed as far as markets go.  I can see retaining him.  Maybe Massie too.  I don't see getting those 3 back

It is fun watching the NFL in general...and the Bears have a somewhat interesting blueprint going on with a rookie QB, rookie HC and a stud defender taking up a ton of cap space...with no picks.  If it works?  Who knows, maybe more teams will go down that road.  There is risk, for sure.  And there is surely reward, too.

I think if the Bears get back to 12 wins?  Well...that should be a tougher 12 wins than it was this year.  Getting to 12 again would be an amazing accomplishment.  

Here's one for yah, Pool.....To get to 12 wins, the magic number, you need to win them all at home and split on the road.  You see that happening?

I think 10 wins is more realistic. I see them taking 4/6 in the division. Winning 3/5 at home (Giants, Cowboys, Chiefs, Chargers, Saints) and 3/4 away (Redskins, Eagles, Rams, Raiders). The thing is though is that what on paper looks like a brutal schedule right now can change drastically. Injuries, (Brees retiring?) etc can happen. Some of those games that look tough now may be easy wins by the time that game rolls around. 

Callahan, Amos, and Massie are all just going to depend on what kind of contracts they want. I'd hate to lose Callahan but if he wants 7m per I think they will let him walk. I think it's almost a foregone conclusion that we lose 2 out of those 3 guys. We do have depth at those positions so they don't necessarily need to draft a rookie and start him though. Bush, McManis, and Kush are all ok. Not great and certainly downgrades but they can hold their own. And there's always hope that Pace can find some more late round gems. He has done really really well with that the last couple of years. 

[dll2000]

I would love to see this injury prediction model. How could you possibly account for all variables?  It would be easier to predict exact  weather 3 weeks from now.  

Regression to mean?  If we had a lot of injuries last year, we won't next year? Or vis a versa? Like saying you got heads 5x in a row so next one will be tails for sure.  Or if you are just taking averages why would one team be any different from another? If that is true what use is it?  What use is it if it can't tell you who would be injured?

My model says each team will suffer, on average, 4 major injuries and 20 medium injuries and 98% of active starters will have minor nagging aches and pains.  Okay great. Now what? Is this going to be an average year, a good year or a bad year? We don't know.  It may take 6 - 10 years for numbers to work to their average.  How does that help me? 

To say Bears didn't have their fair share of key and untimely injuries is false too.  Mack getting hurt cost Miami game. MT getting hurt cost Giants game. Jackson getting hurt cost Philly game.  (Likely for all, who knows what would have happened, but they were all really close games).

Can you cut down on injuries with different practice or training methods or techniques? Of course. That is always the real goal. 

As for Bears next year.  I think they will be a better team on paper, there is no doubt.  Question is will division teams get better too?  It isnt hard to get better when you have a QB in place and every team in division does.   Will opponents be better?  Schedule is defintely harder. Will there be bad injuries to key players? Nobody can know. 

Will Bears win 12 games? Maybe. Will the Packers? 

If i was betting based of info right now i would say Bears are best team in division in 2019. 

[Fl0nkerton]

25 minutes ago, dll2000 said:

Drivel.

Y'all are so BORING.

[pacman5252]

I see chi dropping down 8-9 wins next year. This year they had few major injuries, and a 4th place schedule (didn’t play a playoff team in the road all year). They are tight on cap (20m in space with only 39 guys under contract) and have few picks. For them to get 12 wins with a first place schedule, Mitch would have to improve dramatically, they’d need health luck, and  further in division implosion. 

Im interested to see what the over unders in May will be. I’d guess 8.5-9.5 for the bears, pack, and vikes

[Sasquatch]

2 hours ago, dll2000 said:

I would love to see this injury prediction model. How could you possibly account for all variables?  It would be easier to predict exact  weather 3 weeks from now.  

Regression to mean?  If we had a lot of injuries last year, we won't next year? Or vis a versa? Like saying you got heads 5x in a row so next one will be tails for sure.  Or if you are just taking averages why would one team be any different from another? If that is true what use is it?  What use is it if it can't tell you who would be injured?

My model says each team will suffer, on average, 4 major injuries and 20 medium injuries and 98% of active starters will have minor nagging aches and pains.  Okay great. Now what? Is this going to be an average year, a good year or a bad year? We don't know.  It may take 6 - 10 years for numbers to work to their average.  How does that help me? 

To say Bears didn't have their fair share of key and untimely injuries is false too.  Mack getting hurt cost Miami game. MT getting hurt cost Giants game. Jackson getting hurt cost Philly game.  (Likely for all, who knows what would have happened, but they were all really close games).

Can you cut down on injuries with different practice or training methods or techniques? Of course. That is always the real goal. 

As for Bears next year.  I think they will be a better team on paper, there is no doubt.  Question is will division teams get better too?  It isnt hard to get better when you have a QB in place and every team in division does.   Will opponents be better?  Schedule is defintely harder. Will there be bad injuries to key players? Nobody can know. 

Will Bears win 12 games? Maybe. Will the Packers? 

If i was betting based of info right now i would say Bears are best team in division in 2019. 

Predictive modeling can account for an endless number of variables that assess risk and give you a "probability" with a high level of confidence.  The goal here is to use this information to "optimize" performance, profit, whatever.  So yes, teams would be well advised to do a risk assessment of their team, which would include many variables; average age of the team, # of previous injuries of each player, position of player, injury probability at each player position, etc.  Using this data, they then seek to minimize risk and optimize performance through the methods you mentioned - training techniques, diet, exercise that lessen the probability of injury. 

Regressing to the mean is simple to understand - any random variable that is outside the norm eventually tends to return to the norm.  If the Chicago Bears experienced less injuries in 2018, it's generally expected that the % of injury will return to the norm in subsequent years.  Just look at injury %'s for the Bears over the past 10 years and do the math.  If the injury % for 2018 is extreme (either way), it will be closer to the average next time you measure it in 2019.  In other words, there's a higher probability that the Bear's injury % in 2019 will regress toward the mean. 

I sense you and others take all of this with a grain of salt.  No problem.  In the end it's just sausage making anyways.

[dll2000]

1 hour ago, Sasquatch said:

Predictive modeling can account for an endless number of variables that assess risk and give you a "probability" with a high level of confidence.  The goal here is to use this information to "optimize" performance, profit, whatever.  So yes, teams would be well advised to do a risk assessment of their team, which would include many variables; average age of the team, # of previous injuries of each player, position of player, injury probability at each player position, etc.  Using this data, they then seek to minimize risk and optimize performance through the methods you mentioned - training techniques, diet, exercise that lessen the probability of injury. 

Regressing to the mean is simple to understand - any random variable that is outside the norm eventually tends to return to the norm.  If the Chicago Bears experienced less injuries in 2018, it's generally expected that the % of injury will return to the norm in subsequent years.  Just look at injury %'s for the Bears over the past 10 years and do the math.  If the injury % for 2018 is extreme (either way), it will be closer to the average next time you measure it in 2019.  In other words, there's a higher probability that the Bear's injury % in 2019 will regress toward the mean. 

I sense you and others take all of this with a grain of salt.  No problem.  In the end it's just sausage making anyways.

Risk assessment is fine.  Older players are obvioiusly more likely to get hurt than younger players and recover slower.  Injuries can be cumulative.  Some people are injury prone and some are injury resistant.  Some coaches may produce more injuries through rougher practices or lighter practices that don't prepare you for games depending on one's theory.  Some coaches less.  Or trainers or whoever.  

The part I take with a grain of salt is previous years injuries alone would have any predictive effect whatsoever on subsequent year injuries particularly with different humans.  I am perfectly willing to accept there is an injury mean.  There is a mean for any reoccurring event, but that isn't going to influence one iota the amount of injuries in any given year. 

I used coin example.  But any game of chance could be used.  If 28 comes up 3 times in a row in roulette, odds are still exactly the same it will come up the 4th time or not.  Spin the roulette wheel enough times and it will eventually balance out, but it does not help as a predictor for any given spin (season).  

The mean is useful as a point of reference.  If you are on good side of it you may be just lucky or doing something right.  If you are on bad side of it, you may be unlucky or doing something wrong.  Being on good or bad side of it for several years would be true test if you were doing something wrong or right.  But a coaching and training staff together, would typically only last a few years all at once.   So good data on techniques or unique best practices would be hard to ascertain versus random chance in NFL.

College could be a better place to look since staffs stay together longer, but players are all 18 - 22 and more resistant to injury, recover faster and aren't playing against same size/speed which produces greater risk of injury.

 

 

 

[Sasquatch]

2 minutes ago, dll2000 said:

Risk assessment is fine.  Older players are obvioiusly more likely to get hurt than younger players and recover slower.  Injuries can be cumulative.  Some people are injury prone and some are injury resistant.  Some coaches may produce more injuries through rougher practices or lighter practices that don't prepare you for games depending on one's theory.  Some coaches less.  Or trainers or whoever.  

The part I take with a grain of salt is previous years injuries alone would have any predictive effect whatsoever on subsequent year injuries particularly with different humans.  I am perfectly willing to accept there is an injury mean.  There is a mean for any reoccurring event, but that isn't going to influence one iota the amount of injuries in any given year. 

I used coin example.  But any game of chance could be used.  If 28 comes up 3 times in a row in roulette, odds are still exactly the same it will come up the 4th time or not.  Spin the roulette wheel enough times and it will eventually balance out, but it does not help as a predictor for any given spin (season).  

The mean is useful as a point of reference.  If you are on good side of it you may be just lucky or doing something right.  If you are on bad side of it, you may be unlucky or doing something wrong.  Being on good or bad side of it for several years would be true test if you were doing something wrong or right.  But a coaching and training staff together, would typically only last a few years all at once.   So good data on techniques or unique best practices would be hard to ascertain versus random chance in NFL.

College could be a better place to look since staffs stay together longer, but players are all 18 - 22 and more resistant to injury, recover faster and aren't playing against same size/speed which produces greater risk of injury.

 

 

 

I appreciate that you're trying to work this out in your mind - it's worth continuing for sake of discussion.  I think what you may be missing is that statistically speaking, if you have a season that's extreme (variable that's well off the mean), then it's expected (statistically and mathematically) to move back towards the mean.  It doesn't always happen, but statistically is says it will.  That's why I suggested you look back over the past 10 years at the Bears injury % to see how divergent this past season was from the mean.  If it's extreme, it more likely next year to correct itself.

You bring up all of the variables like new coaches, training practices, etc.  They're all variables that can and should be accounted for when predicting risk and probability.  Predictive modeling is a tool that's used to optimize performance by accounting for as many of the risk variables as possible. 

I disagree that college is any better than any other environment to run predictive modeling - you can place value on any variable - regardless if they're "new or different humans" involved, and you can add as many variables as you like to the model.  It doesn't matter that there's "different human's involved" from year to year, because each of those humans has an r-value that can be quantified and plugged into the formula.  Bottom line is there's an "optimal model" for everything, regardless of the number of variables you introduce to an equation.

What does all of this mean?  Well, statistically speaking, if the Bears had a lower injury % in 2018 than in past years (lower than the mean) than it's a "safe bet" there will be an increase in that % in 2019.  It's also possible to maintain the current low injury % by assessing the current set of variables to try and reproduce optimal results. To do this, you must know the variables, do the math, and engage in activity that's statistically supports optimal performance.  Then you throw the dice.....