NFL Career Lengths by Position (1 Viewer)

[Jeff Pasquino]

FBGs,

I'm looking for a source on the average career length by player position in the NFL.

EX: RB, 3.4 years. QB, 4.7, etc.

Does anyone have one?

TIA.

 

[redman]

FBGs,

I'm looking for a source on the average career length by player position in the NFL.

EX: RB, 3.4 years. QB, 4.7, etc.

Does anyone have one?

TIA.

What good would this information do without a further breakdown based upon the skill of the player, such as draft position or # of starts per year, etc.?

 

[SSOG]

FBGs,

I'm looking for a source on the average career length by player position in the NFL.

EX: RB, 3.4 years. QB, 4.7, etc.

Does anyone have one?

TIA.

What good would this information do without a further breakdown based upon the skill of the player, such as draft position or # of starts per year, etc.?

It would do a lot of good.For instance, let's say that there are 300 WRs, 200 RBs, and 100 TEs in the league. Now, 10% of those WRs will be in the top 10%. 10% of those WRs will be in the bottom 10%. 10% of those RBs will be in the top 10%. 10% of those RBs will be in the bottom 10%. In other words, proportionally speaking, there will be just as many studs at every position, and there will be just as many people just trying desperately to make a 53-man roster. So this means comparisons from position to position will be relatively talent-independent... since proportionally, the talent at RB is distributed exactly like the talent at WR or TE.

If the average career is 3 years for RBs, 5 years for WRs, and 10 years for TEs, well then, that has a lot of implications for Dynasty, even without breaking down by round drafted or whatever. I could use that data to justify drafting TEs higher, since they'd have longer careers, and sluffing the RB position some, since there'd be a lot more churn (which means a lot of new RBs with value every season). I mean, if a TE is getting replaced every 10 seasons, and an RB is getting replaced ever 3, I'd rather have a stud TE since I'll be able to use him longer without replacement (i.e. a stud TE doesn't just get me lots of points, he saves me future draft picks).

Now, I think a case could be made for further breaking it down to compare career lengths of players who make at least one pro bowl in their career (aka the "elite" players) with those who don't make any pro bowls (aka the "replacement level" players). That said, just a general comparison of career lengths by position would still be an incredibly valuable tool without further breakdown.

 

[cstu]

FBGs,

I'm looking for a source on the average career length by player position in the NFL.

EX: RB, 3.4 years. QB, 4.7, etc.

Does anyone have one?

TIA.

What good would this information do without a further breakdown based upon the skill of the player, such as draft position or # of starts per year, etc.?

It would do a lot of good.For instance, let's say that there are 300 WRs, 200 RBs, and 100 TEs in the league. Now, 10% of those WRs will be in the top 10%. 10% of those WRs will be in the bottom 10%. 10% of those RBs will be in the top 10%. 10% of those RBs will be in the bottom 10%. In other words, proportionally speaking, there will be just as many studs at every position, and there will be just as many people just trying desperately to make a 53-man roster. So this means comparisons from position to position will be relatively talent-independent... since proportionally, the talent at RB is distributed exactly like the talent at WR or TE.

If the average career is 3 years for RBs, 5 years for WRs, and 10 years for TEs, well then, that has a lot of implications for Dynasty, even without breaking down by round drafted or whatever. I could use that data to justify drafting TEs higher, since they'd have longer careers, and sluffing the RB position some, since there'd be a lot more churn (which means a lot of new RBs with value every season). I mean, if a TE is getting replaced every 10 seasons, and an RB is getting replaced ever 3, I'd rather have a stud TE since I'll be able to use him longer without replacement (i.e. a stud TE doesn't just get me lots of points, he saves me future draft picks).

Now, I think a case could be made for further breaking it down to compare career lengths of players who make at least one pro bowl in their career (aka the "elite" players) with those who don't make any pro bowls (aka the "replacement level" players). That said, just a general comparison of career lengths by position would still be an incredibly valuable tool without further breakdown.

This is much more useful IMO since most of the players we are concerned with have been to at least one Pro Bowl. It would also be easier to find that info as well.

 

[redman]

SSOG-

The problem is that we don't draft positions, we draft players. I don't decide to draft a TE randomly at a given draft spot just because he has the best chance to play the most years. Most of us don't dare try to project players' careers out more than 2-3 years from the present anyway because there are simply too many variables, e.g coaching changes, free agency, players drafted to challenge their PT, injuries, etc.

We certainly hope that players have long careers, but I have yet to make a draft day decision in dynasty rookie draft based upon how long a player figures to play before retirement. At most I'll look at a guy like Favre and try to extrapolate from other similar players like Elway, Marino, etc. how much longer he figures to play effectively. But using the entire population of players at a position to derive a number seems ludicrous.

Said another way, a system which weights Richie Anderson's 12 seasons in the league with the same weight it gives to Terrell Davis' 7 years has little analytical value.

 

[cstu]

SSOG-

The problem is that we don't draft positions, we draft players. I don't decide to draft a TE randomly at a given draft spot just because he has the best chance to play the most years. Most of us don't dare try to project players' careers out more than 2-3 years from the present anyway because there are simply too many variables, e.g coaching changes, free agency, players drafted to challenge their PT, injuries, etc.

We certainly hope that players have long careers, but I have yet to make a draft day decision in dynasty rookie draft based upon how long a player figures to play before retirement. At most I'll look at a guy like Favre and try to extrapolate from other similar players like Elway, Marino, etc. how much longer he figures to play effectively. But using the entire population of players at a position to derive a number seems ludicrous.

Said another way, a system which weights Richie Anderson's 12 seasons in the league with the same weight it gives to Terrell Davis' 7 years has little analytical value.

That's fine, put some type of minimum on them such as having carried the ball 200 times at least once.

 

[Jeff Pasquino]

FBGs,

I'm looking for a source on the average career length by player position in the NFL.

EX: RB, 3.4 years. QB, 4.7, etc.

Does anyone have one?

TIA.

What good would this information do without a further breakdown based upon the skill of the player, such as draft position or # of starts per year, etc.?

It would do a lot of good.For instance, let's say that there are 300 WRs, 200 RBs, and 100 TEs in the league. Now, 10% of those WRs will be in the top 10%. 10% of those WRs will be in the bottom 10%. 10% of those RBs will be in the top 10%. 10% of those RBs will be in the bottom 10%. In other words, proportionally speaking, there will be just as many studs at every position, and there will be just as many people just trying desperately to make a 53-man roster. So this means comparisons from position to position will be relatively talent-independent... since proportionally, the talent at RB is distributed exactly like the talent at WR or TE.

If the average career is 3 years for RBs, 5 years for WRs, and 10 years for TEs, well then, that has a lot of implications for Dynasty, even without breaking down by round drafted or whatever. I could use that data to justify drafting TEs higher, since they'd have longer careers, and sluffing the RB position some, since there'd be a lot more churn (which means a lot of new RBs with value every season). I mean, if a TE is getting replaced every 10 seasons, and an RB is getting replaced ever 3, I'd rather have a stud TE since I'll be able to use him longer without replacement (i.e. a stud TE doesn't just get me lots of points, he saves me future draft picks).

Now, I think a case could be made for further breaking it down to compare career lengths of players who make at least one pro bowl in their career (aka the "elite" players) with those who don't make any pro bowls (aka the "replacement level" players). That said, just a general comparison of career lengths by position would still be an incredibly valuable tool without further breakdown.

SSOG,You're on the track of where I was going with this - but I need to do more homework on it once I get the info.

Correlating the length of career vs. the "eliteness" of said players is much harder to do, but I'd be willing to state that elite players last longer than non-elites.

However, when you are drafting a rookie - you have no idea if he's going to last 1 year or 10. You don't. Believing 100% contrary to that is fooling yourself. Ryan Leaf is a perfect example (even though he had a 3-year career).

I'm not saying this is the ONLY criteria, but it is important information.

Still Googling... anyone with help let me know.

 

[GregR_2]

If the average career is 3 years for RBs, 5 years for WRs, and 10 years for TEs, well then, that has a lot of implications for Dynasty, even without breaking down by round drafted or whatever. I could use that data to justify drafting TEs higher, since they'd have longer careers, and sluffing the RB position some, since there'd be a lot more churn (which means a lot of new RBs with value every season). I mean, if a TE is getting replaced every 10 seasons, and an RB is getting replaced ever 3, I'd rather have a stud TE since I'll be able to use him longer without replacement (i.e. a stud TE doesn't just get me lots of points, he saves me future draft picks).

If you are going to use the overall numbers, I think you are making a de facto assumption that they have the same general distribution. I wouldn't be surprised either way to find that is or is not the case, as I think I could see arguments to support either way, and I'm not sure which would win out in reality.In short, you don't know if RBs that you would care about for fantasy purposes actually last 6 years but the average is weighted down by a lot of scrubs who get cut in 1... while WRs you might care about might all cluster very tightly around 4 with only a few outliers lower or much higher. If that were the case, you would be making the wrong decision if you went with RB careers being 3 and WRs being 5.

 

[BigTex]

I believe I may have the info your looking for, I'm not sure but I'll see if I can find it when I get home tonight or in the morning but I do remember reading about this on sacbee.com but you have to have a sub. :(

 

[Jeff Pasquino]

I believe I may have the info your looking for, I'm not sure but I'll see if I can find it when I get home tonight or in the morning but I do remember reading about this on sacbee.com but you have to have a sub. :(

Over the years I've opened an online / virtual account with just about every online newspaper, but I appreciate the effort.The information and its usage will become more clear later on, hopefully.

 

[Musesboy]

FBGs,

I'm looking for a source on the average career length by player position in the NFL.

EX:  RB, 3.4 years.  QB, 4.7, etc.

Does anyone have one?

TIA.

What good would this information do without a further breakdown based upon the skill of the player, such as draft position or # of starts per year, etc.?

It would do a lot of good.For instance, let's say that there are 300 WRs, 200 RBs, and 100 TEs in the league. Now, 10% of those WRs will be in the top 10%. 10% of those WRs will be in the bottom 10%. 10% of those RBs will be in the top 10%. 10% of those RBs will be in the bottom 10%. In other words, proportionally speaking, there will be just as many studs at every position, and there will be just as many people just trying desperately to make a 53-man roster. So this means comparisons from position to position will be relatively talent-independent... since proportionally, the talent at RB is distributed exactly like the talent at WR or TE.

If the average career is 3 years for RBs, 5 years for WRs, and 10 years for TEs, well then, that has a lot of implications for Dynasty, even without breaking down by round drafted or whatever. I could use that data to justify drafting TEs higher, since they'd have longer careers, and sluffing the RB position some, since there'd be a lot more churn (which means a lot of new RBs with value every season). I mean, if a TE is getting replaced every 10 seasons, and an RB is getting replaced ever 3, I'd rather have a stud TE since I'll be able to use him longer without replacement (i.e. a stud TE doesn't just get me lots of points, he saves me future draft picks).

Now, I think a case could be made for further breaking it down to compare career lengths of players who make at least one pro bowl in their career (aka the "elite" players) with those who don't make any pro bowls (aka the "replacement level" players). That said, just a general comparison of career lengths by position would still be an incredibly valuable tool without further breakdown.

 

[dgreen]

Can't you just use Drinen's database, sum the number of years for each player (ID), group them by position, then find the average. While not exact, I doubt it's too far from the truth.

 

[dgreen]

I used Drinen's database from 1960-2003 (the latest year I have right now). Of course, this leaves out seasons played before 1960 and after 2003. It also doesn't include any player left out of the database for whatever reason. There are tons and tons of players that never had a single carry, but may have been on a roster. Anyways, maybe it's a start. Here are the means.

QB: 6.36

RB: 3.09

WR: 5.20

TE: 5.64

 

[Jeff Pasquino]

I used Drinen's database from 1960-2003 (the latest year I have right now). Of course, this leaves out seasons played before 1960 and after 2003. It also doesn't include any player left out of the database for whatever reason. There are tons and tons of players that never had a single carry, but may have been on a roster. Anyways, maybe it's a start. Here are the means.

QB: 6.36

RB: 3.09

WR: 5.20

TE: 5.64

dgreen,Thanks.

-Jeff

 

[Jeff Pasquino]

As a follow-up question or questions:

1. Would you agree or disagree that rookie picks are more valuable the more RBs you have to start in your league?

2. Would you agree or disagree that rookie picks are riskier the more RBs you have to start in your league?

Given the short careers of RBs, I'm interested in hearing other opinions.

 

[SSOG]

This is much more useful IMO since most of the players we are concerned with have been to at least one Pro Bowl. It would also be easier to find that info as well.

Funny, I've never been to a rookie dynasty draft where most of the players had been to at least one pro bowl already.

SSOG-

The problem is that we don't draft positions, we draft players. I don't decide to draft a TE randomly at a given draft spot just because he has the best chance to play the most years. Most of us don't dare try to project players' careers out more than 2-3 years from the present anyway because there are simply too many variables, e.g coaching changes, free agency, players drafted to challenge their PT, injuries, etc.

We certainly hope that players have long careers, but I have yet to make a draft day decision in dynasty rookie draft based upon how long a player figures to play before retirement. At most I'll look at a guy like Favre and try to extrapolate from other similar players like Elway, Marino, etc. how much longer he figures to play effectively. But using the entire population of players at a position to derive a number seems ludicrous.

Said another way, a system which weights Richie Anderson's 12 seasons in the league with the same weight it gives to Terrell Davis' 7 years has little analytical value.

I understand the arguement, but I would counter back by saying that this data really only has use in a Dynasty or Keeper league... and in a dynasty league, we really are frequently just drafting a position. If I think that both Reggie Bush and Jay Cutler are going to be studs, then absolutely I let average career length stats sway my decision one way or another.

If you are going to use the overall numbers, I think you are making a de facto assumption that they have the same general distribution. I wouldn't be surprised either way to find that is or is not the case, as I think I could see arguments to support either way, and I'm not sure which would win out in reality.

In short, you don't know if RBs that you would care about for fantasy purposes actually last 6 years but the average is weighted down by a lot of scrubs who get cut in 1... while WRs you might care about might all cluster very tightly around 4 with only a few outliers lower or much higher. If that were the case, you would be making the wrong decision if you went with RB careers being 3 and WRs being 5.

I do understand that problem, and believe that the data is very likely to have biases like the one you mentioned (mostly as a result of special teams, since in my opinion special teamers are more likely to come from some positions- i.e. WRs- than from others- i.e. QBs). Another source of bias is roster makeups. On typical NFL rosters, there are 2 backup QBs for every starter, and 1 or fewer backup DBs for every starter, which means that QBs would tend to have longer careers simply because there's a higher proportion of roster spots devoted to them.That said, it's not like you'll be making any decisions based entirely off of this information- you'll just use it as a tool, combined with a common sense check, to aid in the decision making process.

 

[LHUCKS]

I would be interested to see the info as well.

 

[GregR_2]

As a follow-up question or questions:

1. Would you agree or disagree that rookie picks are more valuable the more RBs you have to start in your league?

2. Would you agree or disagree that rookie picks are riskier the more RBs you have to start in your league?

Given the short careers of RBs, I'm interested in hearing other opinions.

1. The top few rookie picks may become more valuable, but the rest of the rookie picks become less valuable because of #2, that they are riskier.2. Riskier, because the # of successful RBs hasn't changed, but by making other positions less valuable compared to RBs, more picks should probably be spent on RBs.

 

[Jeff Pasquino]

Still looking for good data.

 

[Riffraff]

FBGs,

I'm looking for a source on the average career length by player position in the NFL.

EX: RB, 3.4 years. QB, 4.7, etc.

Does anyone have one?

TIA.

An official study by Dartmouth, has the equations used, hypothesis, and reveals average career spans per position. From 2002.http://www.dartmouth.edu/~chance/chance_ne...1.02.html#item3

position                  expected career length--------------------------------------------------all positions           5.33running backs          4.35wide receivers         4.54defensive tackles      4.71tight ends              4.98cornerbacks            5.13defensive ends         5.39linebackers             5.50safeties                5.88offensive line          5.95quarterbacks            6.96punters/kickers         8.33

 

[Thorpe]

FBGs,

I'm looking for a source on the average career length by player position in the NFL.

EX: RB, 3.4 years. QB, 4.7, etc.

Does anyone have one?

TIA.

An official study by Dartmouth, has the equations used, hypothesis, and reveals average career spans per position. From 2002.http://www.dartmouth.edu/~chance/chance_ne...1.02.html#item3

position                  expected career length--------------------------------------------------all positions           5.33running backs          4.35wide receivers         4.54defensive tackles      4.71tight ends              4.98cornerbacks            5.13defensive ends         5.39linebackers             5.50safeties                5.88offensive line          5.95quarterbacks            6.96punters/kickers         8.33

QBs and O linemen don't see the field as soon as other positions. Safeties and linebackers usually play special teams too, so that helps them stay in the league a little longer.

 

[Doug Drinen]

Here is what I get for players who played in at least one pro bowl. I included all players whose debut year was between 1970 and 1990.

Number that played exactly N years----------------------------------------------------------------------      AVG    1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20----------------------------------------------------------------------qb  13.0    0  0  0  1  0  1  0  1  1  6  3  8  2  4  7  4  4  0  2  0rb   8.6    0  2  1  5  5  8  7 15  7  9  6  6  4  4  0  1  0  0  0  0wr  11.1    0  0  0  0  1  4  4  3  9 10  9  6  7  7  0  5  2  0  0  1te  10.5    0  0  0  1  1  0  3  2  4  3  4  3  6  4  0  0  0  0  0  0ol  12.5    0  0  0  0  0  0  0  5  6  9  6 16 20 11  2  3  3  2  1  1dl  11.5    0  0  0  0  2  1  2  6  6 13  8 10  8  6 10  2  0  0  1  0lb  11.3    0  0  0  0  0  1  3  5  6 12 10 17  9  6  2  2  0  0  1  0db  10.5    0  0  0  1  1  2  6 10 16 15 12  8  6  7  3  2  1  0  0  1

 

[Jeff Pasquino]

Here is what I get for players who played in at least one pro bowl. I included all players whose debut year was between 1970 and 1990.

Number that played exactly N years---------------------------------------------------------------------- AVG 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20----------------------------------------------------------------------qb 13.0 0 0 0 1 0 1 0 1 1 6 3 8 2 4 7 4 4 0 2 0rb 8.6 0 2 1 5 5 8 7 15 7 9 6 6 4 4 0 1 0 0 0 0wr 11.1 0 0 0 0 1 4 4 3 9 10 9 6 7 7 0 5 2 0 0 1te 10.5 0 0 0 1 1 0 3 2 4 3 4 3 6 4 0 0 0 0 0 0ol 12.5 0 0 0 0 0 0 0 5 6 9 6 16 20 11 2 3 3 2 1 1dl 11.5 0 0 0 0 2 1 2 6 6 13 8 10 8 6 10 2 0 0 1 0lb 11.3 0 0 0 0 0 1 3 5 6 12 10 17 9 6 2 2 0 0 1 0db 10.5 0 0 0 1 1 2 6 10 16 15 12 8 6 7 3 2 1 0 0 1

Great info guys.Doug, what does that average out to be?

Could this be done vs. Round 1 selected players?

 

[cstu]

As a follow-up question or questions:

1. Would you agree or disagree that rookie picks are more valuable the more RBs you have to start in your league?

2. Would you agree or disagree that rookie picks are riskier the more RBs you have to start in your league?

Given the short careers of RBs, I'm interested in hearing other opinions.

1. Agree. RB's are a scarce commodity and the more you increase that scarcity the most valuable the rookie RB's will be. 2. As the value of the rookie picks go up, they do become more risky. If people are willing to trade you top WR's and established RB's for say the 1.01, then there's more risk because of the opportunity cost of using the pick rather than trading it. However, RB is the one position where the top rookies continually made an immediate impact so it's a fairly safe position to draft.

 

[cstu]

Here is what I get for players who played in at least one pro bowl. I included all players whose debut year was between 1970 and 1990.

Number that played exactly N years---------------------------------------------------------------------- AVG 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20----------------------------------------------------------------------qb 13.0 0 0 0 1 0 1 0 1 1 6 3 8 2 4 7 4 4 0 2 0rb 8.6 0 2 1 5 5 8 7 15 7 9 6 6 4 4 0 1 0 0 0 0wr 11.1 0 0 0 0 1 4 4 3 9 10 9 6 7 7 0 5 2 0 0 1te 10.5 0 0 0 1 1 0 3 2 4 3 4 3 6 4 0 0 0 0 0 0ol 12.5 0 0 0 0 0 0 0 5 6 9 6 16 20 11 2 3 3 2 1 1dl 11.5 0 0 0 0 2 1 2 6 6 13 8 10 8 6 10 2 0 0 1 0lb 11.3 0 0 0 0 0 1 3 5 6 12 10 17 9 6 2 2 0 0 1 0db 10.5 0 0 0 1 1 2 6 10 16 15 12 8 6 7 3 2 1 0 0 1

That is fantastic. The QB's really stand out for their longevity since only 4 Pro Bowl QB's played less than 10 years. It looks like with Pro Bowl RB's there is about a 50/50 chance of making it to 8 years.

Pro Bowl WR's for the most part make it at least 9 years, more longevity than I expected.

 

[Jeff Pasquino]

Here is what I get for players who played in at least one pro bowl. I included all players whose debut year was between 1970 and 1990.

Number that played exactly N years---------------------------------------------------------------------- AVG 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20----------------------------------------------------------------------qb 13.0 0 0 0 1 0 1 0 1 1 6 3 8 2 4 7 4 4 0 2 0rb 8.6 0 2 1 5 5 8 7 15 7 9 6 6 4 4 0 1 0 0 0 0wr 11.1 0 0 0 0 1 4 4 3 9 10 9 6 7 7 0 5 2 0 0 1te 10.5 0 0 0 1 1 0 3 2 4 3 4 3 6 4 0 0 0 0 0 0ol 12.5 0 0 0 0 0 0 0 5 6 9 6 16 20 11 2 3 3 2 1 1dl 11.5 0 0 0 0 2 1 2 6 6 13 8 10 8 6 10 2 0 0 1 0lb 11.3 0 0 0 0 0 1 3 5 6 12 10 17 9 6 2 2 0 0 1 0db 10.5 0 0 0 1 1 2 6 10 16 15 12 8 6 7 3 2 1 0 0 1

That is fantastic. The QB's really stand out for their longevity since only 4 Pro Bowl QB's played less than 10 years. It looks like with Pro Bowl RB's there is about a 50/50 chance of making it to 8 years.

Pro Bowl WR's for the most part make it at least 9 years, more longevity than I expected.

It isn't as "fantastic" as you think. Players who have achieved the highest of individual honors are given the latitude to stick around longer - hence they will have a longer career.

Would anyone have wanted Jerry Rice over the past few years, either as a player or a fantasy owner?

 

[Jeff Pasquino]

I think the most telling data would come from guys drafted in Round 1, or on Day 1 (Rds 1-3).

 

[Doug Drinen]

I think the most telling data would come from guys drafted in Round 1, or on Day 1 (Rds 1-3).

The draft info section of my database is still very much a work in progress, so take these numbers with a gigantic grain of salt. In fact, I'm absolutely positive that they're not exactly right. But I think they are probably pretty close. Please just ignore them if you're not comfortable with that.But here is what I get for first round picks (between 1970 and 1990):

Number that played exactly N years----------------------------------------------------------------------      AVG    1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20----------------------------------------------------------------------qb   9.8    0  0  5  2  1  1  3  1  3  5  1  5  1  2  4  1  1  0  1  0rb   7.1    2  2 10  7  8  5 14 12  7  3  5  3  3  2  0  1  0  0  0  0wr   8.8    0  1  2  3  7  3  5  4  5  6  2  3  2  1  0  2  2  0  0  1te   8.1    1  0  0  2  3  0  1  2  2  1  0  2  3  0  0  0  0  0  0  0

 

[cstu]

Here is what I get for players who played in at least one pro bowl. I included all players whose debut year was between 1970 and 1990.

Number that played exactly N years---------------------------------------------------------------------- AVG 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20----------------------------------------------------------------------qb 13.0 0 0 0 1 0 1 0 1 1 6 3 8 2 4 7 4 4 0 2 0rb 8.6 0 2 1 5 5 8 7 15 7 9 6 6 4 4 0 1 0 0 0 0wr 11.1 0 0 0 0 1 4 4 3 9 10 9 6 7 7 0 5 2 0 0 1te 10.5 0 0 0 1 1 0 3 2 4 3 4 3 6 4 0 0 0 0 0 0ol 12.5 0 0 0 0 0 0 0 5 6 9 6 16 20 11 2 3 3 2 1 1dl 11.5 0 0 0 0 2 1 2 6 6 13 8 10 8 6 10 2 0 0 1 0lb 11.3 0 0 0 0 0 1 3 5 6 12 10 17 9 6 2 2 0 0 1 0db 10.5 0 0 0 1 1 2 6 10 16 15 12 8 6 7 3 2 1 0 0 1

That is fantastic. The QB's really stand out for their longevity since only 4 Pro Bowl QB's played less than 10 years. It looks like with Pro Bowl RB's there is about a 50/50 chance of making it to 8 years.

Pro Bowl WR's for the most part make it at least 9 years, more longevity than I expected.

It isn't as "fantastic" as you think. Players who have achieved the highest of individual honors are given the latitude to stick around longer - hence they will have a longer career.

Would anyone have wanted Jerry Rice over the past few years, either as a player or a fantasy owner?

Not many guys hang around as long as Rice after their prime. Most players usually hang on a year or two longer than they should. I think the info about Pro Bowl players is more useful that how long 1st round picks stay in the league since you have no idea whether they are backups or starters. If a guy makes a Pro Bowl he usually remains a starter.

 

[redman]

I think the most telling data would come from guys drafted in Round 1, or on Day 1 (Rds 1-3).

The draft info section of my database is still very much a work in progress, so take these numbers with a gigantic grain of salt. In fact, I'm absolutely positive that they're not exactly right. But I think they are probably pretty close. Please just ignore them if you're not comfortable with that.But here is what I get for first round picks (between 1970 and 1990):

                     Number that played exactly N years----------------------------------------------------------------------      AVG    1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20----------------------------------------------------------------------qb   9.8    0  0  5  2  1  1  3  1  3  5  1  5  1  2  4  1  1  0  1  0rb   7.1    2  2 10  7  8  5 14 12  7  3  5  3  3  2  0  1  0  0  0  0wr   8.8    0  1  2  3  7  3  5  4  5  6  2  3  2  1  0  2  2  0  0  1te   8.1    1  0  0  2  3  0  1  2  2  1  0  2  3  0  0  0  0  0  0  0

Great work. That tightened up the variances a bit.

Still, given the fact that a player's fantasy career path tends along some variation of a bell curve, with the zenith reached at different points for different positions, it's awfully tough to draft based even in part upon this information. Is a RB who is going to be a top 10 RB for years 2-6 of his career really worth less on draft day than a QB who will be top ten at his position for years 6-14?

If anything, this is more of a FF free agency/trade strategy tool, which helps you to know when you should rid yourself of players whose best days are behind them.

 

[Jeff Pasquino]

There are two theories working together here, and they are starting to diverge.

1. Pro-Bowl based info. That would be good for FA / cutting a guy after a certain period. However - this is a "backward-looking" number that only comes in to play AFTER the player reaches Pro Bowl status.

2. Draft round info. That would be most appropriate for Dynasty / Rookie drafts. This is "looking foward" numbers - you don't know if he will be a bust or a Pro Bowl caliber player.

I'm looking to work on 2 rather than 1 right now.

 

[Jeff Pasquino]

I think the most telling data would come from guys drafted in Round 1, or on Day 1 (Rds 1-3).

The draft info section of my database is still very much a work in progress, so take these numbers with a gigantic grain of salt. In fact, I'm absolutely positive that they're not exactly right. But I think they are probably pretty close. Please just ignore them if you're not comfortable with that.But here is what I get for first round picks (between 1970 and 1990):

Number that played exactly N years---------------------------------------------------------------------- AVG 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20----------------------------------------------------------------------qb 9.8 0 0 5 2 1 1 3 1 3 5 1 5 1 2 4 1 1 0 1 0rb 7.1 2 2 10 7 8 5 14 12 7 3 5 3 3 2 0 1 0 0 0 0wr 8.8 0 1 2 3 7 3 5 4 5 6 2 3 2 1 0 2 2 0 0 1te 8.1 1 0 0 2 3 0 1 2 2 1 0 2 3 0 0 0 0 0 0 0

Doug, you're the best.Question though - this flies in the face of the average career being just 5.33 years for an average player. Would this seem to indicate that 1st rounders last longer, or are given more latitude to stick around?

 

[Sabertooth]

Is there information more recent than this? I have no idea how to get a good idea of this. I am not savvy enough to get this information out of Drinen's database. Does anyone have a good source for what we can expect as far as career length for top players. I mean obviously crappy players don't last, but what about pretty good guys like Romo, MJD, and Roddy White? What is a reasonable expectation for predicting the sunset of their careers?

 

[Sabertooth]

Found this article.

http://www.rotoworld.com/articles/nfl/40610/277/older-wide-receivers?pg=1

 

[3nOut]

Question though - this flies in the face of the average career being just 5.33 years for an average player. Would this seem to indicate that 1st rounders last longer, or are given more latitude to stick around?

I agree. I never liked hearing when players look for more money since their window is so small. If they were above average, they would have longer careers. If they are below average, then they maybe don't deserve larger contracts.Another question I have is how many of these RBs that have had 7 plus year careers were bell cow RBs and how many of those went to the HOF. I have always wondered what the minimum amount of years a bellcow RB needs to make the HOF. With the NFL moving toward multi-back systems, I wonder how many RBs would prefer to be run to the ground vs. having longer careers (due to splitting time). Would someone rather play 6 years and make the Hall of Fame or play 12 and make more money but no HOF?

 

[Dinsy Ejotuz]

Is there information more recent than this? I have no idea how to get a good idea of this. I am not savvy enough to get this information out of Drinen's database. Does anyone have a good source for what we can expect as far as career length for top players. I mean obviously crappy players don't last, but what about pretty good guys like Romo, MJD, and Roddy White? What is a reasonable expectation for predicting the sunset of their careers?

There are a couple good data points on Doug's chart -- namely the spikes at years 7/8 for RBs, 9-10 for WRs and 8-9 for TEs.Since almost all players enter the league at 21-23 my interpretation is that barring injury most talented RBs tend to play until they're 29-30, talented WRs are good enough to go until 31-32, and talented TEs play until they're 30-31. And the truly elite guys can stretch that. That jibes with other research I've seen and done as well.

 

[Sabertooth]

Is there information more recent than this? I have no idea how to get a good idea of this. I am not savvy enough to get this information out of Drinen's database. Does anyone have a good source for what we can expect as far as career length for top players. I mean obviously crappy players don't last, but what about pretty good guys like Romo, MJD, and Roddy White? What is a reasonable expectation for predicting the sunset of their careers?

There are a couple good data points on Doug's chart -- namely the spikes at years 7/8 for RBs, 9-10 for WRs and 8-9 for TEs.Since almost all players enter the league at 21-23 my interpretation is that barring injury most talented RBs tend to play until they're 29-30, talented WRs are good enough to go until 31-32, and talented TEs play until they're 30-31. And the truly elite guys can stretch that.



That jibes with other research I've seen and done as well.

I speak jibe.

 

[ZWK]

Is there information more recent than this? I have no idea how to get a good idea of this. I am not savvy enough to get this information out of Drinen's database. Does anyone have a good source for what we can expect as far as career length for top players. I mean obviously crappy players don't last, but what about pretty good guys like Romo, MJD, and Roddy White? What is a reasonable expectation for predicting the sunset of their careers?

I did some analysis of RB & WR aging patterns which I described in this post; graph here.Based on the data there, my best guess is that 2014 will be the last fantasy-relevant season for both MJD & Roddy White; perhaps a shade later for White and a shade earlier for Jones-Drew. MJD is a running back coming off a RB1 season at age 26; players meeting that description have had about 2.9 years left until their last season as a fantasy starter (on average). Roddy White is a wide receiver coming off a WR1 season at age 30; players meeting that description have had about 3.1 years left until their last season as a fantasy starter (on average). (And you could adjust that 2.8 for MJD & 3.3 for White, based on their exact age, since MJD has a March birthday and White a November birthday.)

(To read the graph, look at the highest blue line for WRs and the highest orange line for RBs. Look for the number at the bottom that matches the age that the player turned in his last completed season - e.g., 26 for MJD since he turned 26 in 2011. The other lines are based on slightly different definitions of "years left", as described in my other post.)

 

[Truman]

http://www.pro-football-reference.com/blog/?p=6510

couple of weeks ago, I looked at Peter King's comment on Eric Berry to examine whether safeties were risks at the top of the draft. In that post, I determined that safeties were not risky in the sense that they didn't pan out. In fact, compared to other positions, a higher percentage of safeties panned out in that they made at least one pro bowl and were at least good starters (career AV > 49).

However, King did raise a point about safeties being less likely to stay healthy, citing recent injuries to star safeties Troy Polamalu, Bob Sanders, and Ed Reed. Today, I try to take a stab at looking at that issue. The best way we have of doing that is to look at games played data, though games played is also susceptible to talent and ability as well as health. In an attempt to deal with that fact (though I am sure I fail in limited specific cases), I decided to look at players who already proved to be good players at a young age (age 25 and under) and see how many games they played from ages 26 to 29, how frequently they retired before age 30, and how old they were during the last season they were able to play 10 or more games for an NFL team. The hope here is that if a player has played well through age 25, he should continue to play in games through age 29, unless he missed those games due to injury, or was benched or forced to retire due to injury-related decline.

So, I pulled the top fifty players (and ties) as measured by Approximate Value, for all defensive players who turned 25 between 1978 and 2005, at the following positions: defensive tackle/nose tackle, defensive end, inside/middle linebacker, outside linebacker, cornerback, and strong/free safety. A couple of footnotes to that. First, a player had to qualify within one of those positions exclusively. If a player switched positions between, say, outside linebacker and middle linebacker, before he turned 25, then he would have to qualify based only on the points at one of those positions. If he had 25 points of AV at outside linebacker and 5 at inside linebacker, he may have made the list. If it was an even split of 15 at OLB and 15 at ILB, he did not. So, players could have switched positions and still made this list, but it had to be on the strength of only one position. Second, switches within a position did not matter and were counted together--safeties could switch from free safety to strong safety, defensive ends could switch sides, defensive tackles could move to nose tackle. Finally, we always have to balance sample size and sample relevance. I would loved to have a larger sample to examine, but I felt by going much past the top 50 over a span of less than 30 years would bring in guys who were playing because of need/draft position only, and would miss games because of non-injury reasons more frequently.

After coming up with the list of players for each position, I recorded the percentage of games played from ages 26 to 29, and, for retired players, the age at which they last played in 10+ games. Let's get to the results. The columns represent the number of players included in that position grouping, the average percentage of available games played by the young "stars" at that position from ages 26 to 29, the "final age" is the average age that the retired players at that position last played 10+ games, the 33% column represents the percentage of players that missed at least one-third of the available games from ages 26 to 29, and the "done by 30" column represents the percentage of players that played their final season of 10+ games at age 29 or younger.

position no. game pct final age missed 33% done before 30

OLB 51 0.877 31.9 0.098 0.196

CB 52 0.827 31.9 0.173 0.250

DT 54 0.813 31.1 0.148 0.267

DE 51 0.808 31.3 0.235 0.333

ILB 51 0.766 29.9 0.294 0.489

S 50 0.763 29.8 0.260 0.533

Well, look at that. Despite my earlier criticism of King, safety does check in as more injury prone than other defensive positions, in a virtual dead heat with inside linebackers (and my data set does not include any of the three safeties he cited). Just over half of the star young safeties were effectively done or retired before they reached age 30. Compare that to the other secondary position, where 25% of the cornerbacks were done before age 30. The other thing that stands out to me is the size of the difference between playing outside linebacker versus inside linebacker on your life expectancy. I am sure there might be some agents and players interested in seeing how much a switch from the outside to inside will cost them in terms of future career length, as the young star outside linebackers played 2 seasons more on average, and played in almost a half-season's more games from age 26 to age 29. In fact, the outside linebacker who played in the fewest games after turning 26 was Thomas "Hollywood" Henderson, and his demise wasn't because of on-field injuries.

Even though King was proven correct in his assertion that safeties may break down due to being smaller bodies who throw themselves around, I don't think it then follows that drafting safeties early is a bad idea. In fact, using this information, the opposite may be true. As I pointed out in the previous post, a team should be primarily concerned with what a draft pick provides while playing under the initial rookie contract. If we have a position that has a relatively low chance of providing complete busts, and where veteran players in their late twenties tend to fall apart more frequently than other positions, that strikes me as the exact situation when you would want to use your draft picks on a position. We see that on the other side of the ball with running backs, as teams tend to use a high frequency of picks at running back, but are hesitant to give big money long term deals to veterans past a certain age.

 

[coolnerd]

 

[Crazy Tony]

Here is what I get for players who played in at least one pro bowl.  I included all players whose debut year was between 1970 and 1990.

                     Number that played exactly N years----------------------------------------------------------------------      AVG    1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20----------------------------------------------------------------------qb  13.0    0  0  0  1  0  1  0  1  1  6  3  8  2  4  7  4  4  0  2  0rb   8.6    0  2  1  5  5  8  7 15  7  9  6  6  4  4  0  1  0  0  0  0wr  11.1    0  0  0  0  1  4  4  3  9 10  9  6  7  7  0  5  2  0  0  1te  10.5    0  0  0  1  1  0  3  2  4  3  4  3  6  4  0  0  0  0  0  0ol  12.5    0  0  0  0  0  0  0  5  6  9  6 16 20 11  2  3  3  2  1  1dl  11.5    0  0  0  0  2  1  2  6  6 13  8 10  8  6 10  2  0  0  1  0lb  11.3    0  0  0  0  0  1  3  5  6 12 10 17  9  6  2  2  0  0  1  0db  10.5    0  0  0  1  1  2  6 10 16 15 12  8  6  7  3  2  1  0  0  1                                                                                                                                                          <{POST_SNAPBACK}>

That is fantastic. The QB's really stand out for their longevity since only 4 Pro Bowl QB's played less than 10 years. It looks like with Pro Bowl RB's there is about a 50/50 chance of making it to 8 years.

Pro Bowl WR's for the most part make it at least 9 years, more longevity than I expected.

<{POST_SNAPBACK}>

It isn't as "fantastic" as you think. Players who have achieved the highest of individual honors are given the latitude to stick around longer - hence they will have a longer career.

Would anyone have wanted Jerry Rice over the past few years, either as a player or a fantasy owner?

[/QUOTE]I notice in the chart for QBs....the average is 13 years. But there are only 2 QBs in the 13 year category. I think it would be good to know how many starts you got out of a player. If a player is starting they're probably decent.

What about carries for RBs? Starts for RBs?

Jeff,

I remember looking at data from FBGs historical data dominator or just data dominator to compile some stats. Won't that tell you?

For my research I was wanting to know the Top 10 performers in each category for the last 10yrs so I knew who to target potentially.

It was pretty conistent at QB..usual suspects.

A lot of similar names at RB but very rarely would someone stay in there the whole time. window was 5years tops it seemed.

WR very consistent. SHuffling of names sometimes they pop in and out.

Using this information. I figured I would have to replace my stud every 5 years. This means drafting RB every 5 years.

If I focused on WR...it had about a 7 year window.

If I did QBs...well...They don't come around as often and it's harder to predict.

Somewhere in there I gain a 1st, 2nd and 3rd pick. If I can use my 3rd to trade up to a 2nd and 2nds to trade up to a first I'm gaining value.

As long as I don't screw up too badly on my picks I'm OK...or I have more room to make errors

History of picks:

2008 1.04 Matt Forte (2.07 Desean Jackson, 3.04 Jerome Simpson)

2009 1.05 LeSean McCoy (Kenny Britt Josh Freeman) Traded McCoy for Harvin at the draft to Westbrook Owner.

2010 1.05 Hardesty....fail. #1.09 McCluster...Fail. ( other picks Bradford, Gerhart, Edecker, Hernandez)

2011 1.06 G Little (instead of Cam Newton...DAMN), 1.09 DeMarco Murray.) no seconds....no worhtwhile 3rd to mention.

2012 1.02, 2.02, 2.09, 3.02, 3.09.

I don't know if it's true.

I've made some dumb moves but I've been come out ahead on others and pick ups.

Seems to me that if you could determine an average # of starts for a RB or # of carries... Ingram starts but got 10 carries.... vs (insert workhorse)

 

[texasbirdfan]

somewhere along the line about a half a dozen years ago- it was statistically "proven" that the average life of a NFL RB was around 4.5 years. Anyone else remember that? It seems more in line than 7 or 8 years.