Running back Workload vs. Age: A discussion (1 Viewer)

[Chase Stuart]

I've written three articles in an attempt to untangle the messy relationship between workload and age. The first one is here:

http://subscribers.footballguys.com/apps/article.php?article=stuart-running-back-workload

And the next two will be released over the next week.

But I will probably wind up writing a Part IV or even Part V, so I welcome any suggestions you folks may have on additional ways to research the issue.

 

[cstu]

I looked at the best fantasy seasons for RB's over 30 and if workload were really an important factor then there were be more low mileage backs than Priest Holmes and Tiki Barber at the top.

This sums up my thoughts on the subject:

Every carry for a RB is a chance at a major injury so a high workload RB has a higher chance of being injured. However, high workload backs get those carries because they can be trusted to stay healthy.

Age and major injuries are what end RB careers. A low mileage RB might have a better shot at playing well past 30 (Priest Holmes, Barber, Riggins, Thomas Jones) since they've had but less carries and therefore less chance to be injured. However, in the end age and injury catch up to everyone regardless of mileage.

 

[Bracie Smathers]

I've written three articles in an attempt to untangle the messy relationship between workload and age. The first one is here:

http://subscribers.footballguys.com/apps/article.php?article=stuart-running-back-workload

And the next two will be released over the next week.

But I will probably wind up writing a Part IV or even Part V, so I welcome any suggestions you folks may have on additional ways to research the issue.

Hey Chase,

Looking for the correlation between age and workload you get right to the point and focus-in on attempts since that is basically what you're trying to tie-in with the age of a RB and you use historic data to try and get a picture and come to a conclusion but their is something recently that seems to be a definite trend that would effect any conclusions you come to based on historic data.

I think it is obvious the league has shifted away from the running game on a whole but within the lower overall rushing attempts stemming from, rules favoring receivers, the new recieiving TEs, the spread offenses with multiple WRs/TEs, etc. In addition to that we have seen OCs have conciously gone away from the workhorse RBs to full blown RBBCs.

The recent trend would limit rush attempts for all RBs and that would apply to RBs who make it to the ages you are focusing-in on with your articles.

So my suggestion would be to weight the historical data lessor than the more recent data where lower overall rushing attempts and RBBC would seem to have a differnt set of data that might draw different conclusions.

We might see some guys like, oh a Toby Garheart who has low mileage/attempts and is approaching the ages you are interested in and where he might get classic workhorse attempts. IOWs the new RBs who make it to older ages might be like a Toby Garheart, guys who haven't been used like the bulk of the RBs who came from a differnt era of high rush attempts that are highlighted in your article.

 

[Adam Harstad]

Hey Chase, nice work, and I look forward to reading the rest.

When I've looked at issues like this in the past where I'm looking to compare two groups at the opposite end of a broad spectrum (e.g. high-mileage vs. low-mileage RBs, speedy vs. slow WRs, etc), I use a method I cribbed from one of Doug's old posts. Basically, I establish a certain criteria, find all players who met that criteria, sort those players based on where they fall on that spectrum, then compare the top third to the bottom third.

I think where Doug went wrong when he first used that method is that he set the baselines too high. He came up with a list of all RBs with four top-12 finishes through age 27 and compared the "low-mileage" to the "high-mileage" backs from that group. That's not a conundrum most fantasy owners face regularly- if you have four top-12 finishes by age 27, then pretty much by definition you're not a low-mileage back. I think most fantasy owners are more concerned with a Michael Turner-type situation, where he had 228 career carries up to age 26, and then was the #2 fantasy RB at age 26. How would Turner age compared to, say, a Maurice Jones-Drew, who had 1151 carries up to age 26, and then was the #3 fantasy RB at age 26? How would a Larry Johnson (140 carries prior to age 26) age compared to LaDainian Tomlinson (1363 carries prior to age 26).

If I were to take a look at it, that'd probably be my method- take all backs to finish in the top 10 at age 26 (or 27, or whatever), label the top third in prior workload as "high mileage", the bottom third as "low mileage", and use the middle third as a control, and then see who had more career remaining afterwards. In my mind, that would be the best way to answer the question that most fantasy owners REALLY mean to ask when they ask whether age is more important than mileage- the question of whether this recent stud with light career usage will play longer than this other recent stud with heavy career usage.

 

[KellysHeroes]

I have no data, just what I have seen. The only RB I'm interested in having after age 28 are the ultra talented ones with low little injuries like LT and Peterson. Their the only ones that seem to produce at age 30 and a little beyond. And the best part is you can get them cheap at age 28, the price you pay is well worth the FF points you'll recieve for the next 3 yrs.

 

[Dinsy Ejotuz]

Just messed with this myself by looking at backs who did well from age 27-29 (I forget the criteria, but they're all good backs) and then looking at average games played from 30+.

Certainly doesn't look like usage before 27 matters at all.

 

[KellysHeroes]

gotta remember, ages 25 - 30 is usually an adults' prime years.

 

[JaxBill]

Just for arguments's sake have you considered taking total touches into account.

i.e Jamaal Charles attempts were down slightly last year to 259 but his receptions doubled to 70.

On a different note, what worries me are RB who have a huge workload in a given year(s). MJD was a case in point. Lightly used behind Fred taylor, then had 3 years with a huge number of touches. Injury cut the fourth year short. He had a heavy workload last year but obviously isn't the same back. Is it the injury or is it the workload that led up to the injury?

 

[jurb26]

Just messed with this myself by looking at backs who did well from age 27-29 (I forget the criteria, but they're all good backs) and then looking at average games played from 30+.

Certainly doesn't look like usage before 27 matters at all.

This is how I see it as well. Late 20s pounding can add up. Mid to early 20s, the body is just so resilient it's unreal.

 

[Adam Harstad]

Just messed with this myself by looking at backs who did well from age 27-29 (I forget the criteria, but they're all good backs) and then looking at average games played from 30+.

Certainly doesn't look like usage before 27 matters at all.

I've looked at it several different times in several different ways. I've also read multiple articles on the subject on the old PFR blog and by Chase at Football Perspective. This is always the conclusion I've reached, too. Every single look has either determined that the effect of workload was so small as to be meaningless, so weak as to be statistically insignificant, or in some cases it has determined that prior workload has a positive effect on remaining career length.

It's important to remember exactly what question we're asking, though. The question isn't "if we had an identical clone of Adrian Peterson who had 1000 fewer career carries, would we expect him to have a longer career in front of him?" The answer to that question is probably "yes". It's also completely irrelevant unless and until someone makes an identical clone of Adrian Peterson.

The real question that we are asking is "if we have two backs who seem identical, but one of whom has more career carries, which one would we expect to last longer", and in this case, the negative effect of all of those career carries is offset by the fact that carries serve as an indicator of player quality. Lots of carries indicate a better player, and better players last longer. Adrian Peterson with 1500 career carries might be more worn down than, say, Michael Turner with 300 career carries... but he's also a better player, which is why he got those 1500 carries in the first place. A worn-down Adrian Peterson might be expected to be roughly equivalent to a fresh Michael Turner.

Also playing a role is the fact that carries are not distributed randomly. Coaches decide who get carries. They make this decision in part based on their opinion of who can HANDLE those carries. In this case, if a guy doesn't have a lot of career carries, that might mean he's fresher... but it also might serve as an indicator that his previous coaches have not believed he could handle a large workload.

Either way, in the real world, I have yet to see any compelling evidence suggesting workload matters in the slightest bit when predicting decline. When I'm evaluating dynasty RBs, age is really the only variable I care about.

 

[Adam Harstad]

On a different note, what worries me are RB who have a huge workload in a given year(s). MJD was a case in point. Lightly used behind Fred taylor, then had 3 years with a huge number of touches. Injury cut the fourth year short. He had a heavy workload last year but obviously isn't the same back. Is it the injury or is it the workload that led up to the injury?

This seems like a Texas Sharpshooter fallacy to me. Maurice Jones-Drew had, by your own admission, THREE huge workload seasons (after already adding a not-insignificant number of touches in his first three seasons). If he'd broken down after his first huge workload season, we'd say "wow, that workload really did him in". He didn't- instead, he actually improved his per-game yardage totals (though he did miss two games) and got an even bigger workload. If he'd broken down after that second season, we'd have said his two-year workload did him in. He didn't- instead, he put up arguably his best pro season, leading the league in rushing for a bad Jacksonville franchise. And then when he broke down the next year, we said his workload did him in and he hasn't been the same back since. But if workload is the problem, why wasn't it a problem after his first two huge workload seasons?

Every back that has a large workload will break down eventually- not necessarily because of workload, but because father time is undefeated. For each of those backs, we can go back after the fact and say "wow, he had a huge workload leading up to that, I'm sure that contributed". But that's an after-the-fact observation; during the middle of his high-workload run, we really have no idea when he's going to go over the cliff.

That's why strictly-defined tests like these are necessary. If we're defining everything after the fact, we're going to see in the data what we want or expect to see. By defining success and failure before we ever look at the data, we help mitigate that risk and get an objective answer.

 

[JaxBill]

On a different note, what worries me are RB who have a huge workload in a given year(s). MJD was a case in point. Lightly used behind Fred taylor, then had 3 years with a huge number of touches. Injury cut the fourth year short. He had a heavy workload last year but obviously isn't the same back. Is it the injury or is it the workload that led up to the injury?

This seems like a Texas Sharpshooter fallacy to me. Maurice Jones-Drew had, by your own admission, THREE huge workload seasons (after already adding a not-insignificant number of touches in his first three seasons). If he'd broken down after his first huge workload season, we'd say "wow, that workload really did him in". He didn't- instead, he actually improved his per-game yardage totals (though he did miss two games) and got an even bigger workload. If he'd broken down after that second season, we'd have said his two-year workload did him in. He didn't- instead, he put up arguably his best pro season, leading the league in rushing for a bad Jacksonville franchise. And then when he broke down the next year, we said his workload did him in and he hasn't been the same back since. But if workload is the problem, why wasn't it a problem after his first two huge workload seasons?

Every back that has a large workload will break down eventually- not necessarily because of workload, but because father time is undefeated. For each of those backs, we can go back after the fact and say "wow, he had a huge workload leading up to that, I'm sure that contributed". But that's an after-the-fact observation; during the middle of his high-workload run, we really have no idea when he's going to go over the cliff.

That's why strictly-defined tests like these are necessary. If we're defining everything after the fact, we're going to see in the data what we want or expect to see. By defining success and failure before we ever look at the data, we help mitigate that risk and get an objective answer.

Yeah, I get that and is the reason why I asked it as a question.

I guess the better question to me is do you look at a player like MJD differently if his touches (for simpler math) by season go 100, 150, 300, 350, 400?

After season 5 do you treat him as a 27 year-old that had 1400 touches over 5 years or as a guy who had 1050 in last 3 years?

 

[Snotbubbles]

Take a look at the Stephen Jackson thread from last year. There was a lot of discussion about workload and age.

 

[pantherclub]

Take a look at the Stephen Jackson thread from last year. There was a lot of discussion about workload and age.

http://forums.footballguys.com/forum/index.php?showtopic=679240&page=12&hl=%2Bsteven+%2Bjackson#entry16667546

Lots of good info in there but there are also some quality fails as well. some people refuse to believe.

 

[CalBear]

I have no data, just what I have seen. The only RB I'm interested in having after age 28 are the ultra talented ones with low little injuries like LT and Peterson. Their the only ones that seem to produce at age 30 and a little beyond. And the best part is you can get them cheap at age 28, the price you pay is well worth the FF points you'll recieve for the next 3 yrs.

A blown ACL is a little injury?

 

[KellysHeroes]

I have no data, just what I have seen. The only RB I'm interested in having after age 28 are the ultra talented ones with low little injuries like LT and Peterson. Their the only ones that seem to produce at age 30 and a little beyond. And the best part is you can get them cheap at age 28, the price you pay is well worth the FF points you'll recieve for the next 3 yrs.

A blown ACL is a little injury?

these days, I would consider 1 a hiccup yes... anymore than 1 and its time to get nervous

 

[12punch]

there could be all kinds of correlations between all this stuff, and I'm not against taking closer looks at anything, but it seems to me all these kinds of conversations have the underlying flaw of bias --- people have already imagined that rb get worn down through use, so they set out to basically find this result.

look at the type of analogies that get freely thrown around --- 'tread getting worn off tires', or 'putting mileage on' for example.

why use that -- why phrase it like that?

people are clearly not cars, and legs are clearly not tires, but in the poster's imagination they've already decided the results are similar for no particular reason, so they pick this image to reinforce something they've already dreamed up and now want to give substance.

when you work with your hands your skin toughens and thickens --- it doesn't wear through like the interior upholstery of a car might.

when you go to the weight room every other day are you putting mileage on your muscles and wearing them out like a motor, or are you strengthening them through a biological process?

are marathoners taking tread off their legs or are they building endurance when they run countless miles to train for a race?

jerry rice grew up catching bricks --- was that putting mileage on his hands and wearing the tread off his fingers to shorten his career?

people are not cars, they are biological organisms.

another problem I see is that all these numbers are completely arbitrary --- touches and carries are determined by the arbitrary framework of nfl rules, not by some scientific study of human endurance.

if we were talking about this back in the 40s via telegraph, or whatever, we'd be using 200 carries as the 'heavy usage' number instead of 300 because the season was only 10 games -- now 200 is considered the light usage guy.

how about when they extend the season?

initially, I'll bet you see coaches using more rbbc because they've become entrenched in the 16 game mentality, but 5 yrs in do you think matt forte still gets capped at 360 touches, or will all these guys just start breaking down every other year because they constantly cross this magic threshold of 300 touches, or 350, or 400, or whatever it's currently set at?

what about playoff games --- are we adding in those touches, or they don't count?

 

[12punch]

I have no data, just what I have seen. The only RB I'm interested in having after age 28 are the ultra talented ones with low little injuries like LT and Peterson. Their the only ones that seem to produce at age 30 and a little beyond. And the best part is you can get them cheap at age 28, the price you pay is well worth the FF points you'll recieve for the next 3 yrs.

A blown ACL is a little injury?

I can't speak for the guy, but if peterson has 1 blown acl then that's clearly a low number of blown acls, if that's what the guy meant, and the guy never claimed it was a little injury --- that's you doing that.

in 2014 one blown acl is probably preferable in the long term to repeated chronic stuff like ankle sprains, hammies, and messed up knees.

 

[CalBear]

I have no data, just what I have seen. The only RB I'm interested in having after age 28 are the ultra talented ones with low little injuries like LT and Peterson. Their the only ones that seem to produce at age 30 and a little beyond. And the best part is you can get them cheap at age 28, the price you pay is well worth the FF points you'll recieve for the next 3 yrs.

A blown ACL is a little injury?

I can't speak for the guy, but if peterson has 1 blown acl then that's clearly a low number of blown acls, if that's what the guy meant, and the guy never claimed it was a little injury --- that's you doing that.

in 2014 one blown acl is probably preferable in the long term to repeated chronic stuff like ankle sprains, hammies, and messed up knees.

Are you aware that the ACL is a component of the knee joint?

Certainly sports science has gotten better at repairing and rehabbing ACL injuries. But it's among the most major surgeries an athlete is likely to endure. Not only is the ACL itself gone, but a piece of tendon is removed to replace it; both the ligament and the tendon need to heal and are more subject to future injury.

The only "low" number of blown ACLs is zero.

 

[12punch]

I have no data, just what I have seen. The only RB I'm interested in having after age 28 are the ultra talented ones with low little injuries like LT and Peterson. Their the only ones that seem to produce at age 30 and a little beyond. And the best part is you can get them cheap at age 28, the price you pay is well worth the FF points you'll recieve for the next 3 yrs.

A blown ACL is a little injury?

I can't speak for the guy, but if peterson has 1 blown acl then that's clearly a low number of blown acls, if that's what the guy meant, and the guy never claimed it was a little injury --- that's you doing that.

in 2014 one blown acl is probably preferable in the long term to repeated chronic stuff like ankle sprains, hammies, and messed up knees.

Are you aware that the ACL is a component of the knee joint?

Certainly sports science has gotten better at repairing and rehabbing ACL injuries. But it's among the most major surgeries an athlete is likely to endure. Not only is the ACL itself gone, but a piece of tendon is removed to replace it; both the ligament and the tendon need to heal and are more subject to future injury.

The only "low" number of blown ACLs is zero.

maybe I got you confused with somebody else, but I had thought you were some kind of statistician, so I'd give you credit for being a little brighter than this, unless you're just deliberately trolling for argument, as seems to be so popular on this board.

I'll try to explain it again, but if you didn't pick it up the first 2 times I doubt the third will get much traction.

this time I have bolded 2 different words to draw your attention --- those words are completely different in meaning.

again, I can't speak for somebody else, but nowhere in his post do I see him referring to a blown acl as a 'little' injury.

that's something you came up with in your own head and are apparently dead set on proving yourself wrong for some nutty reason.

 

[sn0mm1s]

I have no data, just what I have seen. The only RB I'm interested in having after age 28 are the ultra talented ones with low little injuries like LT and Peterson. Their the only ones that seem to produce at age 30 and a little beyond. And the best part is you can get them cheap at age 28, the price you pay is well worth the FF points you'll recieve for the next 3 yrs.

A blown ACL is a little injury?

I can't speak for the guy, but if peterson has 1 blown acl then that's clearly a low number of blown acls, if that's what the guy meant, and the guy never claimed it was a little injury --- that's you doing that.

in 2014 one blown acl is probably preferable in the long term to repeated chronic stuff like ankle sprains, hammies, and messed up knees.

Are you aware that the ACL is a component of the knee joint?

Certainly sports science has gotten better at repairing and rehabbing ACL injuries. But it's among the most major surgeries an athlete is likely to endure. Not only is the ACL itself gone, but a piece of tendon is removed to replace it; both the ligament and the tendon need to heal and are more subject to future injury.

The only "low" number of blown ACLs is zero.

maybe I got you confused with somebody else, but I had thought you were some kind of statistician, so I'd give you credit for being a little brighter than this, unless you're just deliberately trolling for argument, as seems to be so popular on this board.

I'll try to explain it again, but if you didn't pick it up the first 2 times I doubt the third will get much traction.

this time I have bolded 2 different words to draw your attention --- those words are completely different in meaning.

again, I can't speak for somebody else, but nowhere in his post do I see him referring to a blown acl as a 'little' injury.

that's something you came up with in your own head and are apparently dead set on proving yourself wrong for some nutty reason.

What? I read that exactly the same way as CalBear. I don't see how he isn't referring to a blown ACL as a little injury since he is classifying all injuries that AD and LT sustained as "little". I mean, in the response he said a blown ACL is a "hiccup". I would say you are the one trolling here.

 

[Shutout]

I have been part of these types of discussions, in depth, literally, at least 50 times over the years with many people that built some well-thought out designs on how to measure/predict this.

In my own system, I built a measurer/predictor and found that Curtis Martin absolutely destroyed my math no matter how I tweaked it. So I labeled him as the alpha-outlier and tossed him out. I also seemed to hit the nail on the head, literally within weeks of when Eddie George and Shaun Alexander would break. So I was thinking I was on to something. Then I had misses, hits, etc. In general, random. I predicted SJAX's demise and was about 9 months off.

So, my conclusion: without a crystal ball, this is one of the most fruitless exercises the FF community can participate in because:

-NONE of us have put anything together yet that works in two decades.

-AND...the data is now more of a moving target than ever. We always used to pull the emmitt Smiths, Fred Taylors, Larry Johnsons, Marshall Faulks, etc, out and use them as measuring sticks. Now the NFL has changed and nobody is being used like LT and ADP, etc.

It is more of a crapshoot than ever before and is a complete waste of time because of these two reasons:

1-IF anyone ever figures it out, it will be the single biggest advantage that has ever existed in FF.

2-Because of #1, no one will be able to keep it to themselves and, within a season or so, EVERYONE will have the exact same information....which is where we are now in the media/info age...which means all the great advantages to FF are all but gone. You can't walk into a live draft anymore, pick a guy named Carl Pickens, get laughed at, and then win your league because he has a pro-bowl year. You can, but if you do, its just the same dumb luck where we all find that nut every once in a while. You can't hold that advantage consistently anymore.

So, while its a nice discussion, it is irrelevant to how we should spend our time in FF these days.

 

[Tau837]

I have no data, just what I have seen. The only RB I'm interested in having after age 28 are the ultra talented ones with low little injuries like LT and Peterson. Their the only ones that seem to produce at age 30 and a little beyond. And the best part is you can get them cheap at age 28, the price you pay is well worth the FF points you'll recieve for the next 3 yrs.

A blown ACL is a little injury?

I can't speak for the guy, but if peterson has 1 blown acl then that's clearly a low number of blown acls, if that's what the guy meant, and the guy never claimed it was a little injury --- that's you doing that.

in 2014 one blown acl is probably preferable in the long term to repeated chronic stuff like ankle sprains, hammies, and messed up knees.

Are you aware that the ACL is a component of the knee joint?

Certainly sports science has gotten better at repairing and rehabbing ACL injuries. But it's among the most major surgeries an athlete is likely to endure. Not only is the ACL itself gone, but a piece of tendon is removed to replace it; both the ligament and the tendon need to heal and are more subject to future injury.

The only "low" number of blown ACLs is zero.

maybe I got you confused with somebody else, but I had thought you were some kind of statistician, so I'd give you credit for being a little brighter than this, unless you're just deliberately trolling for argument, as seems to be so popular on this board.

I'll try to explain it again, but if you didn't pick it up the first 2 times I doubt the third will get much traction.

this time I have bolded 2 different words to draw your attention --- those words are completely different in meaning.

again, I can't speak for somebody else, but nowhere in his post do I see him referring to a blown acl as a 'little' injury.

that's something you came up with in your own head and are apparently dead set on proving yourself wrong for some nutty reason.

What? I read that exactly the same way as CalBear. I don't see how he isn't referring to a blown ACL as a little injury since he is classifying all injuries that AD and LT sustained as "little". I mean, in the response he said a blown ACL is a "hiccup". I would say you are the one trolling here.

 

[Adam Harstad]

I have been part of these types of discussions, in depth, literally, at least 50 times over the years with many people that built some well-thought out designs on how to measure/predict this.

In my own system, I built a measurer/predictor and found that Curtis Martin absolutely destroyed my math no matter how I tweaked it. So I labeled him as the alpha-outlier and tossed him out. I also seemed to hit the nail on the head, literally within weeks of when Eddie George and Shaun Alexander would break. So I was thinking I was on to something. Then I had misses, hits, etc. In general, random. I predicted SJAX's demise and was about 9 months off.

So, my conclusion: without a crystal ball, this is one of the most fruitless exercises the FF community can participate in because:

-NONE of us have put anything together yet that works in two decades.

-AND...the data is now more of a moving target than ever. We always used to pull the emmitt Smiths, Fred Taylors, Larry Johnsons, Marshall Faulks, etc, out and use them as measuring sticks. Now the NFL has changed and nobody is being used like LT and ADP, etc.

It is more of a crapshoot than ever before and is a complete waste of time because of these two reasons:

1-IF anyone ever figures it out, it will be the single biggest advantage that has ever existed in FF.

2-Because of #1, no one will be able to keep it to themselves and, within a season or so, EVERYONE will have the exact same information....which is where we are now in the media/info age...which means all the great advantages to FF are all but gone. You can't walk into a live draft anymore, pick a guy named Carl Pickens, get laughed at, and then win your league because he has a pro-bowl year. You can, but if you do, its just the same dumb luck where we all find that nut every once in a while. You can't hold that advantage consistently anymore.

So, while its a nice discussion, it is irrelevant to how we should spend our time in FF these days.

I don't necessarily agree with this. I think the current fantasy market is pricing older running backs as if previous workload is a useful component in predicting decline. If it turns out that workload is NOT a useful component in predicting decline, then that means the current fantasy market is being inefficient in its pricing of high-workload RBs, and that inefficiency can be exploited for a profit.

Or, to put it another way, if conventional wisdom says we should downgrade RBs with a ton of career carries, and actual results suggest that we should not, then we can make a relatively easy profit simply by betting against conventional wisdom.

 

[dickey moe]

What do you all think of the chances that the following have productive careers into their 30s?:

1) Peterson

2) Forte

3) Lynch

4) CJohnson

 

[pantherclub]

What do you all think of the chances that the following have productive careers into their 30s?:

1) Peterson

2) Forte

3) Lynch

4) CJohnson

define productive

 

[josefrees]

This is really ancillary aspect of the issue and hard to quantify but there could be positive aspects of blown knees. It is most likely media posturing but more and more guys are saying they are more explosive post injury, once the recovery process is complete. And another hard to quantify context, does a year off due to injury (for a guy besides AP for whom a torn ACL may well be a little hiccup) help longevity?

This is a question that is close to my interests because of the Eagles backfield this year. Sproles and Shady have nearly the same amount of career touches yet Sproles has been in the league for twice as long. Who has more wear?

PS. I think that high school and college workload are an underrated and often ignored aspect of a guys career. It isn't like your counter starts over again once you get drafted. Some guys have been half ran into the ground before they even hit a professional team.

 

[12punch]

. I predicted SJAX's demise and was about 9 months off.

is sjax officially demised?

as a guy who drafted him last year, I'm painfully aware of his 2013 season, but can we safely attribute that to age and not other factors?

personally, I think I'm going back to the well if he slides --- which you'd think he would.

 

[jurb26]

. I predicted SJAX's demise and was about 9 months off.

is sjax officially demised?

as a guy who drafted him last year, I'm painfully aware of his 2013 season, but can we safely attribute that to age and not other factors?

personally, I think I'm going back to the well if he slides --- which you'd think he would.

Jackson is a shell of his glory days but that Atl offense was a mess last year. He should perform much better so long as they stay healthy.

 

[pantherclub]

. I predicted SJAX's demise and was about 9 months off.

is sjax officially demised?

as a guy who drafted him last year, I'm painfully aware of his 2013 season, but can we safely attribute that to age and not other factors?

personally, I think I'm going back to the well if he slides --- which you'd think he would.

Jackson is a shell of his glory days but that Atl offense was a mess last year.

. I predicted SJAX's demise and was about 9 months off.

is sjax officially demised?

as a guy who drafted him last year, I'm painfully aware of his 2013 season, but can we safely attribute that to age and not other factors?

personally, I think I'm going back to the well if he slides --- which you'd think he would.

yes. Age and his career workload are the tell tale signs

He had 2 maybe 3 average games last year. Thats after getting hurt early on, as many on here predicted based on his prior years yet people still want this headache on their team. As I said in the other thread, let someone else deal with him he is not worth a roster space.

 

[12punch]

. I predicted SJAX's demise and was about 9 months off.

is sjax officially demised?

as a guy who drafted him last year, I'm painfully aware of his 2013 season, but can we safely attribute that to age and not other factors?

personally, I think I'm going back to the well if he slides --- which you'd think he would.

Jackson is a shell of his glory days but that Atl offense was a mess last year.

. I predicted SJAX's demise and was about 9 months off.

is sjax officially demised?

as a guy who drafted him last year, I'm painfully aware of his 2013 season, but can we safely attribute that to age and not other factors?

personally, I think I'm going back to the well if he slides --- which you'd think he would.

yes. Age and his career workload are the tell tale signs

He had 2 maybe 3 average games last year. Thats after getting hurt early on, as many on here predicted based on his prior years yet people still want this headache on their team. As I said in the other thread, let someone else deal with him he is not worth a roster space.

that's fair enough -- I certainly don't have all the answers.

but if his injuries are predictable, and his production was due to age, then we can safely predict another significant injury for him in 2014, and his numbers should be even worse, as he's aged another year --- am I correct?

 

[pantherclub]

/\ yes

I would bet anything that he gets hurt again this year. Its certainly been the trend

****edit before people go nuts here - I am not saying you can predict all injuries just with certain players with the history, workload, age and oline worries it makes it really easy when these benchmarks are all hit

 

[CalBear]

This is really ancillary aspect of the issue and hard to quantify but there could be positive aspects of blown knees. It is most likely media posturing but more and more guys are saying they are more explosive post injury, once the recovery process is complete. And another hard to quantify context, does a year off due to injury (for a guy besides AP for whom a torn ACL may well be a little hiccup) help longevity?

I guarantee you won't find any doctor or trainer who thinks that blowing out your knee could improve your longevity.

 

[Biabreakable]

Here is a study I came across a couple of weeks ago on the subject I found interesting.

This one focuses on running back performance in fantasy football by age and workload.

This is a study I found last year about player performance after an ACL injury.

The last one has been debunked somewhat I think by Adrian Peterson and Jamal Charles. So back to the drawing board there. However the observations about a younger players recovery being more likely than an older player likely has some truth to it. I also found the part about players who are the highest level performers being more likely to have this type of injury to be pretty interesting.

 

[ZWK]

I'd like to see more about how receiving fits into things. When I looked at some numbers a while back, it looked like RBs who are good receivers tend to age better than those who aren't active in the receiving game. But I didn't take a very rigorous look at it. I also recall FO claiming that a drop in receiving production from an older RB is often a sign of imminent decline.

Jason Lisk had some posts a few years back on the PFR blog looking at workload by game (rather than by season, or by career). I believe that he found that extremely high carry games were predictive of injuries. I'd be interested in seeing more analysis along those lines, both because 1) I remember it being among the stronger evidence pointing to a negative effect of workload and 2) the theory that high-carry games increase risk sounds plausible even if high-carry seasons (or careers) don't have an effect (or are too confounded by coaches' decisions for it to be possible to find an effect in the data).

 

[Dinsy Ejotuz]

Imagine a normal curve. The vertical axis is RB production and the max potential height and width of the curve varies by talent level. The horizontal axis is age.

Now imagine that every RB's potential career is represented by a horizontal line cutting across that curve. The better the RB the lower the talent line intersects across the curve.

So great RBs are capable of not only producing more, sooner, but also capable of producing later. All else (esp injuries) being equal.

The problem with all these age-related studies is that they try to treat every RB's curve the same, or they don't account adequately for talent level of the back in question. ZWK's study got around some of that by looking at VBD produced, but even that can be misleading. Rudi Johnson (228 VBD thru age 27) just isn't going to play as long as Reggie Bush (162 VBD thru 29) on average -- because Reggie Bush is a better NFL RB.

IMO the only thing that really helps much beyond the obvious stuff is to try and ID a back's true level of talent and work that into the equation.

 

[dickey moe]

What do you all think of the chances that the following have productive careers into their 30s?:

1) Peterson

2) Forte

3) Lynch

4) CJohnson

define productive

Good question. For the purposes of this discussion, how about: are they worth riding out in dynasty as they enter their 30s?

 

[pantherclub]

What do you all think of the chances that the following have productive careers into their 30s?:

1) Peterson

2) Forte

3) Lynch

4) CJohnson

define productive

Good question. For the purposes of this discussion, how about: are they worth riding out in dynasty as they enter their 30s?

Probably Peterson but at some point he is going to hit the wall sooner rather than later.

 

[cloppbeast]

there could be all kinds of correlations between all this stuff, and I'm not against taking closer looks at anything, but it seems to me all these kinds of conversations have the underlying flaw of bias --- people have already imagined that rb get worn down through use, so they set out to basically find this result.

look at the type of analogies that get freely thrown around --- 'tread getting worn off tires', or 'putting mileage on' for example.

why use that -- why phrase it like that?

people are clearly not cars, and legs are clearly not tires, but in the poster's imagination they've already decided the results are similar for no particular reason, so they pick this image to reinforce something they've already dreamed up and now want to give substance.

when you work with your hands your skin toughens and thickens --- it doesn't wear through like the interior upholstery of a car might.

when you go to the weight room every other day are you putting mileage on your muscles and wearing them out like a motor, or are you strengthening them through a biological process?

are marathoners taking tread off their legs or are they building endurance when they run countless miles to train for a race?

jerry rice grew up catching bricks --- was that putting mileage on his hands and wearing the tread off his fingers to shorten his career?

people are not cars, they are biological organisms.

another problem I see is that all these numbers are completely arbitrary --- touches and carries are determined by the arbitrary framework of nfl rules, not by some scientific study of human endurance.

if we were talking about this back in the 40s via telegraph, or whatever, we'd be using 200 carries as the 'heavy usage' number instead of 300 because the season was only 10 games -- now 200 is considered the light usage guy.

how about when they extend the season?

initially, I'll bet you see coaches using more rbbc because they've become entrenched in the 16 game mentality, but 5 yrs in do you think matt forte still gets capped at 360 touches, or will all these guys just start breaking down every other year because they constantly cross this magic threshold of 300 touches, or 350, or 400, or whatever it's currently set at?

what about playoff games --- are we adding in those touches, or they don't count?

Good post.

I don't see how participating in more football could possibly reduce a players ability over time. In no other sport would we apply the same reasoning. 'He's played a lot of horse-shoes in his day, he's bound to lose it soon' - I don't think I'll ever hear that sentence.

Not trying to disparage anyone's effort in this inquiry, but correlation involves finding some functional relationship between data - usually linearly. Arbitrarily setting and threshold introduces bias, and really doesn't tell us much anyway.

 

[Dinsy Ejotuz]

In general I think all really good backs (and those guys on your list are all good) are worth riding into the sunset with. Especially the guys who are good pass catchers and might be able to keep a 3rd down/RB3 role after they fade.

 

[sn0mm1s]

What do you all think of the chances that the following have productive careers into their 30s?:

1) Peterson

2) Forte

3) Lynch

4) CJohnson

define productive

Good question. For the purposes of this discussion, how about: are they worth riding out in dynasty as they enter their 30s?

Peterson is the only one I would take a risk with assuming I could get value for the others.

 

[ZWK]

I wonder what the state of the art is for this kind of analysis in other sports, e.g. pitch counts in baseball. Have analytics guys in other sports come up with methods that we could steal?

 

[Proust Loves Cake]

Just messed with this myself by looking at backs who did well from age 27-29 (I forget the criteria, but they're all good backs) and then looking at average games played from 30+.

Certainly doesn't look like usage before 27 matters at all.

I've looked at it several different times in several different ways. I've also read multiple articles on the subject on the old PFR blog and by Chase at Football Perspective. This is always the conclusion I've reached, too. Every single look has either determined that the effect of workload was so small as to be meaningless, so weak as to be statistically insignificant, or in some cases it has determined that prior workload has a positive effect on remaining career length.

If I could quibble with wording here, wouldn't you agree that those findings suggest that prior workload, if anything, has a "positive association/i]" with career length, rather than a "positive effect"? This is at the heart of the correlation-causation problem. Certainly possible (if not likely) there is an underlying x-factor that drives both early workload and career length.

Just my .02.

 

[Proust Loves Cake]

This is really ancillary aspect of the issue and hard to quantify but there could be positive aspects of blown knees. It is most likely media posturing but more and more guys are saying they are more explosive post injury, once the recovery process is complete. And another hard to quantify context, does a year off due to injury (for a guy besides AP for whom a torn ACL may well be a little hiccup) help longevity?

I guarantee you won't find any doctor or trainer who thinks that blowing out your knee could improve your longevity.

Ordinarily, I'd agree with you on this. However, do some doctors take the position that the integrity of the ACL and/or other ligament tears can be just as intact after a successful surgery as it was prior to injury? I believe I have heard his before.

If that's the case, it doesn't seem entirely unreasonable that a year of ACL tear rehabilitation regimen could be, in some ways, superior than a typical offseason training (e.g., increased motivation, less wear-tear, more efficient training/strengthening of key of muscle groups, etc.). In this way, I don't think it's out of the realm of possibility that a person with a blown out knee could legitimately come back better than ever.

 

[12punch]

This is really ancillary aspect of the issue and hard to quantify but there could be positive aspects of blown knees. It is most likely media posturing but more and more guys are saying they are more explosive post injury, once the recovery process is complete. And another hard to quantify context, does a year off due to injury (for a guy besides AP for whom a torn ACL may well be a little hiccup) help longevity?

I guarantee you won't find any doctor or trainer who thinks that blowing out your knee could improve your longevity.

Ordinarily, I'd agree with you on this. However, do some doctors take the position that the integrity of the ACL and/or other ligament tears can be just as intact after a successful surgery as it was prior to injury? I believe I have heard his before.

If that's the case, it doesn't seem entirely unreasonable that a year of ACL tear rehabilitation regimen could be, in some ways, superior than a typical offseason training (e.g., increased motivation, less wear-tear, more efficient training/strengthening of key of muscle groups, etc.). In this way, I don't think it's out of the realm of possibility that a person with a blown out knee could legitimately come back better than ever.

I've actually heard the repaired acl is stronger and more resistant to future injury, but it's not anything I've really looked into --- you'd want to get bramel's input on something like that, if he wanted to chime in on it.

and your correction on correlation/causality is probably actually more than just a quibble on semantics --- those 2 things get interchanged way too much in common usage to the point of confusion on a lot of things.

 

[CalBear]

This is really ancillary aspect of the issue and hard to quantify but there could be positive aspects of blown knees. It is most likely media posturing but more and more guys are saying they are more explosive post injury, once the recovery process is complete. And another hard to quantify context, does a year off due to injury (for a guy besides AP for whom a torn ACL may well be a little hiccup) help longevity?

I guarantee you won't find any doctor or trainer who thinks that blowing out your knee could improve your longevity.

Ordinarily, I'd agree with you on this. However, do some doctors take the position that the integrity of the ACL and/or other ligament tears can be just as intact after a successful surgery as it was prior to injury? I believe I have heard his before.

If that's the case, it doesn't seem entirely unreasonable that a year of ACL tear rehabilitation regimen could be, in some ways, superior than a typical offseason training (e.g., increased motivation, less wear-tear, more efficient training/strengthening of key of muscle groups, etc.). In this way, I don't think it's out of the realm of possibility that a person with a blown out knee could legitimately come back better than ever.

I've actually heard the repaired acl is stronger and more resistant to future injury, but it's not anything I've really looked into --- you'd want to get bramel's input on something like that, if he wanted to chime in on it.

and your correction on correlation/causality is probably actually more than just a quibble on semantics --- those 2 things get interchanged way too much in common usage to the point of confusion on a lot of things.

Surgically-repaired ACLs are not stronger than healthy ACLs. They may not even be stronger than knees with blown ACLs which have been rehabbed without surgery.

 

[ldizzle]

I always liked this study. I havent taken the time to evaluate it recently though.

http://www.footballdocs.com/running_back_carries.html

 

[Adam Harstad]

I'd like to see more about how receiving fits into things. When I looked at some numbers a while back, it looked like RBs who are good receivers tend to age better than those who aren't active in the receiving game. But I didn't take a very rigorous look at it. I also recall FO claiming that a drop in receiving production from an older RB is often a sign of imminent decline.

Jason Lisk had some posts a few years back on the PFR blog looking at workload by game (rather than by season, or by career). I believe that he found that extremely high carry games were predictive of injuries. I'd be interested in seeing more analysis along those lines, both because 1) I remember it being among the stronger evidence pointing to a negative effect of workload and 2) the theory that high-carry games increase risk sounds plausible even if high-carry seasons (or careers) don't have an effect (or are too confounded by coaches' decisions for it to be possible to find an effect in the data).

I also remember that Lisk study, and thought that was perhaps the most compelling workload-related evidence I'd seen. I do think it might be possible to overwork an RB in a single game and wear him out in the short term. Anecdotally, that seems to be what happened to Knowshon Moreno after the New England game last year.

I wonder how useful analysis of receiving backs would be. Totally anecdotally, it seems like the RBs that have gotten most heavily-involved in the passing game tend to be more talented, on average, than the RBs that haven't. That would certainly confound the data.

A lot of the age vs. workload analysis I see fails to control for one factor when addressing the other. For instance, most studies that find that RBs with a higher workload tend to decline fail to account for the fact that RBs with a higher workload are also, on average, older. Are they declining because of the higher workload, or are they declining because they're older? If you want to answer that, you really need to compare RBs with different workloads at the same age. I think the studies that take that extra step are the ones that tend to find that workload is overrated.

 

[Adam Harstad]

Just messed with this myself by looking at backs who did well from age 27-29 (I forget the criteria, but they're all good backs) and then looking at average games played from 30+.

Certainly doesn't look like usage before 27 matters at all.

I've looked at it several different times in several different ways. I've also read multiple articles on the subject on the old PFR blog and by Chase at Football Perspective. This is always the conclusion I've reached, too. Every single look has either determined that the effect of workload was so small as to be meaningless, so weak as to be statistically insignificant, or in some cases it has determined that prior workload has a positive effect on remaining career length.

If I could quibble with wording here, wouldn't you agree that those findings suggest that prior workload, if anything, has a "positive association/i]" with career length, rather than a "positive effect"? This is at the heart of the correlation-causation problem. Certainly possible (if not likely) there is an underlying x-factor that drives both early workload and career length.Just my .02.

I would absolutely agree with that. I think that gets to the heart of the question that fantasy owners are really asking. When they ask if workload is bad for an RB, they aren't asking "if I could create a clone of Adrian Peterson with 1000 fewer career carries, would he last longer than the original version of ADP?" In that case, I think he probably would. But the real question that faces fantasy owners is "Who will have the longer career between these two different but superficially similar RBs, the one with the high workload or the one with the low workload?" And in that case, the answer is often "the one with the high workload".

 

[Adam Harstad]

I always liked this study. I havent taken the time to evaluate it recently though.

http://www.footballdocs.com/running_back_carries.html

Does anyone know why this analysis always uses 370 carries as the cutoff? I can shed some light on it. Here's every RB in history with 360-369 carries, along with where they ranked in year N and in year N+1:

Adrian Peterson '08 - 3rd that year, 2nd the next

Edgerrin James '05 - 5th that year, 20th the next... but in fairness, he switched from the Colts to the Cards.

Rudi Johnson '04 - 8th that year, 8th the next

Edgerrin James '99 - 2nd that year, 2nd the next

Curtis Martin '99 - 8th that year, 8th the next

Curtis Martin '98 - 8th that year, 8th the next (note: not a typo, Curtis ranked 8th in three consecutive seasons)

Terrell Davis '97 - 2nd that year, 1st the next

Curtis Martin '95 - 2nd that year, 4th the next

Emmitt Smith '94 - 1st that year, 1st the next

Emmitt Smith '91 - 3rd that year, 1st the next

Herschel Walker '88 - 3rd that year, 11th the next (traded at midseason from Dallas to Minnesota)

James Wilder '85 - 8th that year, 28th the next

Earl Campbell '81 - 12th that year, 26th the next

Earl Campbell '79 - 2nd that year, 2nd the next

Walter Payton '79 - 1st that year, 4th the next

Basically, the group of backs with 360-369 carries is one of the most elite, historically-productive cohorts in NFL history. A spectacular 80% managed to resist regression and perform as a fantasy RB1 in the follow-up season, which is an almost unheard-of success rate. 40% (6 of 15) ranked in the top two at their position the following season. So the key takeaway here is that if an RB gets 369 carries in one season, he's going to have one of the best years of his entire career the next year... but if he gets one more carry and crosses that magical 370 mark, he's going to fall off a cliff.

The "curse of 370" is a fantastic example of the rule of multiple endpoints. If you come up with a hypothesis and then look at the data before you select your endpoints, there's a good chance you'll be able to find some endpoints that make the data support your hypothesis.