The Myth of the 'Rookie Wall' (1 Viewer)

[Dinsy Ejotuz]

Just like the myth of the 'injury prone' player, even a cursory glance at the data totally debunks the idea that there's a 'rookie wall' for RBs.

Eight rookie RBs have had 250+ touches* since 2002 (when the FBG database starts including half-season splits):

Joseph Addai

Cadillac Williams

Kevin Jones

Willis McGahee

Domanick Davis

William Green

Clinton Portis

Domanic Rhodes

--All eight of these RBs got more touches in games 9+ than they did in games 1-8

--Average touches among the eight jumped from 14 in the 1st half to 23 in the 2nd half

--Seven of the eight RBs saw their points/touch rise during the 2nd half* of the season (Davis dropped from .64 to .62)

--Average FF points/touch for all eight backs rose 19% from the first half of the season to the second -- .58 to .69 (non-PPR)

The 'rookie wall' is a myth. Rookie RBs get stronger as the season goes on.

* Note that the FBG 2nd-half splits include playoff games.

 

[KnowledgeReignsSupreme]

You're basing this on a sample size of 8?

 

[Dinsy Ejotuz]

Feel free to expand on my research, but it won't change the results. The odds that there's actually a rookie wall, but ALL of the first eight players you look at defy the reality is pretty close to zero.

 

[David Yudkin]

I think the concept of hitting the rookie wall is for players that started the season and played extensively and then had issues or ran out of gas.

Since in today's era many rookies don't get a lot of playing time early on, that makes evaluating the concept pretty difficult. Many of the players you mentioned did not have a full time workload to begin with:

Joseph Addai - had only one game of 20 carries in the first half of the season

Cadillac Williams - got hurt and missed several game early to mid-season

Kevin Jones - had no games with 20 carries the first part of the season

Willis McGahee - barely played the first 5 games (and techically was not a rookie)

Domanick Davis - really only started playing full time in Week 7

William Green - didn't start until Week 11

Clinton Portis - first game of over 20 carries was Week 8

Domanic Rhodes - barely played until Week 8

It's hard to monitor long term effects over a season when all of these guys did not end up playing any more total games than they did in college.

 

[Dinsy Ejotuz]

Thanks David. The fact that no one's had that sort of workload in the last five years is a totally different way to debunk the myth. Either way, it's not worth worrying about.

Does anyone have access to split info further back than 2002?

 

[Dinsy Ejotuz]

Tomlinson, Lewis, James and Robert Edwards all logged 320+ touches between 1997 and 2001, but I can't find game splits.

I'm also wondering if yards per carry is a better way to measure the 1st half vs 2nd half since variations in TDs could muck things up.

 

[David Yudkin]

Thanks David. The fact that no one's had that sort of workload in the last five years is a totally different way to debunk the myth. Either way, it's not worth worrying about.Does anyone have access to split info further back than 2002?

Profootballreference.com has game data back to 1996, but there are no splits to look up . . . the totals would have to calculated by hand. I don't have time now or I would consider doing it.The only rookies in that time that had 300+ carries were Edge, LT, Eddie George, Jamal Lewis, and Karim Abdul-Jabbar.It's rare that a rookie back would be tabbed the clear starter right from the start of the season.As an example of a player that may have hit the rookie wall, Laurence Maroney seemed to run out of gas as the season went on (and injury was likely the culprit).

 

[Doug Drinen]

Here is a newsgroup post I wrote on the topic back in October of 2001 in response to a debate about whether LT might hit the wall.

FWIW.

 

[Dinsy Ejotuz]

I'll do some of this by hand at some point. I'm really curious now whether there's an actual effect.

btw...for the entire NFL yards per rush has risen from weeks 9-17 compared to weeks 1-8 -- so that's something that has to be discounted as well.

2006 4.30 vs 4.21

2005 4.27 vs 4.14

2004 4.18 vs 4.13

2003 4.18 vs 4.11

 

[FantasyTrader]

It would be interesting to see how strength of schedule may have impacted this. 8 RB's isn't a very big sample size. Good research though.

 

[Dinsy Ejotuz]

Thanks Doug...

"Of the 19, 10 did better "post-wall" and 9 did better "pre-wall" (I'm defining "better" in terms of fantasy points per game). As you'll see when you peruse the data below, there's a whole lot of noise and really not enough info to draw too many conclusions.

But if you like anecdotal evidence, consider the three most heavily-used rookie RBs of the past 6 years: Eddie George, Curtis Martin and Edge James. These are the three who are clearly most comparable to Tomlinson from a productivity standpoint. All three finished the season stronger than they started it."

I'm also not convinced that 14 touches/game jumping to 23 over the 2nd half of the season isn't enough of a workload to see a dropoff if there was going to be one. 23 touches per game over half a season is heavy for a kid that's still trying to get acclimated to the NFL game.

 

[FreeBaGeL]

Cool idea for a thread, but a couple things seem to hurt the credibility of the way the study was performed.

250 touches as the main cutoff point kind of comes out of nowhere. As Yudkin pointed out, many of these guys played little early on. If the idea is that the rookie wall comes after a full 11 or 12 games (presumably when the college season would have been over), then many of the guys listed here wouldn't have even made it to that "wall" yet. Some of the guys struggle the first couple games while they get their bearings as well, which hurts these average numbers on their splits.

Also, splitting it between 1st and 2nd half of the season is another arbitrary cutoff that doesn't seem to go along with the "theory". I think 8 games to hit the wall is less than most seem to assume (especially given that the first 8 the player is generally used less). I was really interested to see how according to the numbers Caddy was fine when most of us thought he very obviously hit a wall his rookie year.

I wonder if a better example (if we could get a large enough sample size, which we really can't) would be comparing games 4 or 5 through 11 or 12 (presumably after they've gotten adjusted some and are ready to go full stride) to games 12+ or 13+, where the college football season would have already been over. I don't think anyone is claiming that the "rookie wall" comes in week 9, especially when many of the guys being looked at just fell into the starting roll a couple weeks before. The study here seems to be claiming that the "rookie wall" for some guys would be after 2 or 3 starts, which no one is claiming.

 

[Dinsy Ejotuz]

You could be right, Free. I didn't spend much time on it because the results looked so lopsided to me. The biggest problem is the time to get the data you want.

But at this point b/w what I did and Drinen's link I'd have to say there's enough to suggest that there's no rookie wall, and nothing (so far) to suggest that there's anything to the idea. Addai certainly didn't hit one in the post-season last year.

Also...the last quarter of the season accounts for HALF the 2nd half stats that I posted. A 19% bump in 2nd half points/touch probably wouldn't be possible if half of the 2nd half involved a rookie hitting the wall.

 

[Dinsy Ejotuz]

Going back a bit further one finds there were nine rookie RBs between 1993 and 2001 who logged at least twenty touches per game (320+ for the season).

LaDainian Tomlinson

Jamal Lewis

Edgerrin James

Robert Edwards

Eddie George

Karim Abdul-Jabbar

Curtis Martin

Marshall Faulk

Jerome Bettis

Unfortunately game logs at Pro Football Reference start in 1995, so I don't have information for Faulk and Bettis. Luckily, the following further confirms the initial idea. Grouping the seven rookie seasons together by week you see the following:

Game.....Rush...Yards...Avg

Game 1...146....558....3.82

Game 2...133....475....3.57

Game 3...123....582....4.73

Game 4...122....505....4.14

Game 5...124....554....4.47

Game 6...151....435....2.88

Game 7...160....522....3.26

Game 8...126....459....3.64

Game 9...148....648....4.38

Game 10..153...682....4.46

Game 11..131...439....3.35

Game 12..150...711....4.74

Game 13..174...680....3.91

Game 14..156...711....4.56

Game 15..179...747....4.17

Game 16..141...527....3.74

So...

1st half: 136 average rushes per player and 3.77 YPC

2nd half: 176 average rushes per player and 4.18 YPC

More carries and more yards in the 2nd half of the season. As was true for the players selected in the initial post.

Yards per carry by quarter season?

1st four games, 4.05 YPC

2nd four games, 3.51 YPC

3rd four games, 4.26 YPC

4th four games, 4.10 YPC

There is no such thing as a rookie wall -- outside the superstitious mythology of the Fantasy Football player, that is. Good rookies get more carries, and do more with those carries as their rookie seasons progress.

 

[Thorpe]

Going back a bit further one finds there were nine rookie RBs between 1993 and 2001 who logged at least twenty touches per game (320+ for the season).LaDainian TomlinsonJamal LewisEdgerrin JamesRobert EdwardsEddie GeorgeKarim Abdul-JabbarCurtis MartinMarshall FaulkJerome BettisUnfortunately game logs at Pro Football Reference start in 1995, so I don't have information for Faulk and Bettis. Luckily, the following further confirms the initial idea. Grouping the seven rookie seasons together by week you see the following:Game.....Rush...Yards...AvgGame 1...146....558....3.82Game 2...133....475....3.57Game 3...123....582....4.73Game 4...122....505....4.14Game 5...124....554....4.47Game 6...151....435....2.88Game 7...160....522....3.26Game 8...126....459....3.64Game 9...148....648....4.38Game 10..153...682....4.46Game 11..131...439....3.35Game 12..150...711....4.74Game 13..174...680....3.91Game 14..156...711....4.56Game 15..179...747....4.17Game 16..141...527....3.74So...1st half: 136 average rushes per player and 3.77 YPC 2nd half: 176 average rushes per player and 4.18 YPCMore carries and more yards in the 2nd half of the season. As was true for the players selected in the initial post.Yards per carry by quarter season?1st four games, 4.05 YPC2nd four games, 3.51 YPC3rd four games, 4.26 YPC4th four games, 4.10 YPCThere is no such thing as a rookie wall -- outside the superstitious mythology of the Fantasy Football player, that is. Good rookies get more carries, and do more with those carries as their rookie seasons progress.

How do you explain the 1.6 YPC difference between Game 5 and 6?

 

[Dinsy Ejotuz]

How do you explain the 1.6 YPC difference between Game 5 and 6?

I don't. For any single week the numbers will fluctuate a bit with only seven players. Could be a couple guys faced all-world defenses that week or something.

 

[Chase Stuart]

http://apps.footballguys.com/04stuart_rookiepoints.cfm

 

[dgreen]

Yards per carry by quarter season?1st four games, 4.05 YPC2nd four games, 3.51 YPC3rd four games, 4.26 YPC4th four games, 4.10 YPC

The dreaded 2nd four games wall!

 

[Dinsy Ejotuz]

SSOG's nice post this morning on QBs got me thinking about a couple threads I did a few years ago, so I looked them up. I thought it might be interesting to update this one and see how heavily-used rookie RBs the last three years fared.

There were four new RBs who met the previously-used criteria of 250 carries: Forte, Lynch, Slaton and Chris Johnson (Peterson had 238).

Based on the previous data there's no such thing as a rookie wall, so we'd expect that those four RBs' Y/C and points/touch held steady or increased as the season went on.

And, voila:

Games 1-4: 4.23 yards/carry ...1.26 PPT

Games 5-8: 4.03 yards/carry....1.28 PPT

Games 9-12: 4.52 yards/carry....1.22 PPT

Games 13-16: 4.58 yards/carry...1.25 PPT

Rookie Wall = Still a Myth

 

[Sigmund Bloom]

This is going to be a very difficult thing to prove with numbers, but I don't think the idea that it takes a lot more for a player to survive the rigors of an NFL season than it does to survive a college season is a myth, in fact I think it's an objective fact.

Just looking at 2008 (2009 didnt have any rookies with heavy workloads), Matt Forte most certainly wore down in December, call it the "rookie wall" or not, it happened. Chris Johnson, Kevin Smith, and Steve Slaton didn't, and they weren't ridden as hard as Forte was in September/October. The idea of "the rookie wall" might have helped you save your bacon if you traded Forte based on it.

Adrian Peterson clearly wasn't the same back in the second half of the year as he was in the first half in 2007. Some of that was possibly due to lingering issues from his injury, but it still remains that he just wasn't the same back in the 2nd half of his rookie year, and seemed to run out of gas in December.

Looking back at LT, three of his five worst games in ypc and three of his four worst games in rushing yards came in the last four games of his rookie year. Looks like the rookie wall to me.

I just think that there are going to be too few RBs that get enough work in their rookie year to truly test the rookie wall theory to make any "one size fits all" observations. It sure felt like Forte, Peterson, and LT were being overused in their rookie seasons. Peterson and LT have ended up performing much better in late season games during subsequent seasons. We'll see about Forte. I would not abandon the idea that rookie RBs who have heavy workloads will "hit a wall" in December.

 

[Dinsy Ejotuz]

Sig, how about showing some data to support your assertions and opinions. I've gone to the trouble of proving my case using all available RBs who received a heavy workload in their rookie years. Just saying "I don't think it's true" on the basis of a handful of cherry-picked anecdotes really isn't very useful.

 

[Sigmund Bloom]

Sig, how about showing some data to support your assertions and opinions. I've gone to the trouble of proving my case using all available RBs who received a heavy workload in their rookie years. Just saying "I don't think it's true" on the basis of a handful of cherry-picked anecdotes really isn't very useful.

The sample size of rookie RBs who were workhorses over the first 12 games of their career is just not large enough to conclude anything meaningful from the numbers. I find it much more useful to recollect the RBs who were used as workhorses in recent memory and recall how their Decembers went. I think throwing together a bunch of very different cases and averaging the numbers glosses over reality in an attempt to create a pattern to match, and the NFL in general is about exceptional cases and small sample sizes, it isn't like baseball where there are so many games played every year that everything averages out in the long run.

 

[Dinsy Ejotuz]

Sig, how about showing some data to support your assertions and opinions. I've gone to the trouble of proving my case using all available RBs who received a heavy workload in their rookie years. Just saying "I don't think it's true" on the basis of a handful of cherry-picked anecdotes really isn't very useful.

The sample size of rookie RBs who were workhorses over the first 12 games of their career is just not large enough to conclude anything meaningful from the numbers. I find it much more useful to recollect the RBs who were used as workhorses in recent memory and recall how their Decembers went. I think throwing together a bunch of very different cases and averaging the numbers glosses over reality in an attempt to create a pattern to match, and the NFL in general is about exceptional cases and small sample sizes, it isn't like baseball where there are so many games played every year that everything averages out in the long run.

Well, we'll have to agree to disagree then. I'll take the data over your recollections. If only two or three players in twelve years meet the conditions you put forward there's effectively no wall in any event.

 

[Red Herring]

Is it possible that the "rookie wall" was once a real issue, but with modern conditioning, etc., rookies are better able to push past the ~12 games they're used to? My first year was LT2's as well and I drafted him (thanks to FBG) and later traded (thanks again FBG) him before he would "hit the wall." Hit it he did, with a noticable dropoff in points the last 1/4 of the season. Lately though, playing in dynasty I've had more rookies to watch, and it has seemed that there were fewer "walls" hit as time went on. Now, it is rare to see someone wear down, usually when the rookie held out or was injured or otherwise not able to join in the team's conditioning program ASAP.

Maybe the thing to watch for is 1st year players that don't make it to mini-camps or miss time during preseason in some way? (etaOr perhaps the wall to look for isn't games played, but # of carries, catches, or throws a player was used to in college?

 

[Sigmund Bloom]

Here's an excerpt from 2009 rookie TE James Casey's diary:

On a separate matter, you always hear about the rookie wall. When I was playing in college, I heard a lot of commentators talking about it, and I think I'm kind of experiencing it a little bit right now. Not that it's affected me too much, but your body definitely starts to tell you that you're going through an NFL season about this time of the year (December). A normal college season is only 12 games, and we've already played 12. That's not even counting the four preseason games, which, to a guy like me, were pretty much regular games because I was playing a lot. So that's 16 games, four more than you usually play in a whole college season.

On top of that, right when I was done with my bowl game against Western Michigan here in the Texas Bowl, I realized I wanted to declare for the draft. And right after that, literally a couple of days later, I started training for the combine as hard as I could. After that, I was training for my Pro Day, and then I was training for individual workouts with teams and then we had OTAs. So I've been going nonstop since last January, which is pretty much a full year of no down time at all for the rookies.



It's something you've just got to fight through. I've really been watching all the veterans and seeing what they do to take care of their body. I'm starting to pick up on some stuff like getting to the cold tub and making sure you keep on working out even though you're really tired, just to keep your body in shape. Our strength coaches and athletic staff, we have the best in the world at what they do, so they definitely stay on top of us and make sure we're staying healthy and doing all of the preventative stuff in the weight room.

I know this is just one anecdotal case, and far from proof, but I would imagine that any player would acknowledge the "rookie wall" when speaking candidly (players won't use the rookie wall as an excuse for a dropoff in performance if they have an attitude of full accountability). I just can't believe that it is a myth that an NFL season is harder on a player than a college season, and that the rookie year includes an adjustment to this reality.

 

[Chase Stuart]

I think both Bloom and wdcrob are right in this thread. There's no doubt that, on average, rookie RBs (at least good ones) tend to play better as the season goes on. I've looked at this several times before. Someone like Portis is a good example, as he got more carries as the season went on and he was an uberstud. That said, it's also obvious that some rookie RBs do wear down and aren't used to the heavy workload. I agree that, as Bloom put it, it's difficult to make a "one size fits all" theory about rookie running backs.

That said, predicting which rookie RBs will wear down in the future seems like a sucker bet. Most rookie RBs won't wear down, and I don't think there's a particular set of facts or circumstances (such as a high workload in September) that portends doom in December. There may be -- I might look at the issue a bit more myself since it's been a few years since I fully researched it-- but I don't think that's the case.

I also think that at least some rookie RBs that perform poorly down the stretch aren't hitting a rookie wall but are just having bad games. "Splits happen", and it's important not to simply assign a reason when there in fact is no reason at all. If you re-ordered the season into games 4, 5, 15, 1, 8, 10, 16, etc., several RBs would look like uberstuds in the first half of the season and total crap in the final quarter. But it obviously wouldn't mean that they "broke down" or hit a wall, but rather, it was just the normal random variance associated with a very small sample. Undoubtedly some of the RBs people think hit a rookie wall did no such thing.

 

[Dinsy Ejotuz]

I know this is just one anecdotal case, and far from proof, but I would imagine that any player would acknowledge the "rookie wall" when speaking candidly (players won't use the rookie wall as an excuse for a dropoff in performance if they have an attitude of full accountability). I just can't believe that it is a myth that an NFL season is harder on a player than a college season, and that the rookie year includes an adjustment to this reality.

This makes perfect sense Sigmund. I don't doubt that it's rough for these guys mentally and physically. I'm just saying that as a predictor of performance the rookie wall is a myth. There's no evidence for it at this point.

 

[CalBear]

You see stats without context. It's a bit baffling. Then again, I guess Chase Stuart could use an apprentice.

You list six reasons why a rookie wall could exist. I could list more than six reasons why Bigfoot could exist. If Bigfoot actually exists, his presence is measurable and documentable. We should be able to see Bigfoots alive in the wild, and find their carcasses after they die. The fact that every serious analysis of the data that's been conducted has found no meaningful evidence that Bigfoot exists suggest strongly that he doesn't.No matter how you slice the data, it looks like rookie RBs perform as well or better in the second half of the season than in the first half. To add another data point, I looked at all rookie RBs with 50+ carries in a half-season since 1995. There were more RBs in the second half group (95 vs. 75), they had higher YPC, and on a per-RB basis, the second half group got more carries, more receptions, and more TDs.

 

[az_prof]

You lost me as soon as you said "myth of injury prone players." I am not at all convinced it is a "myth," and I see a lot of evidence to support it. Of course how you define "injury" I suppose and what would constitute "overcoming" the injury, and how many and what type of injuries constitute "injury prone" are all subjective. But I have been watching football for forty years and I know that some guys seem to be constantly getting hurt and end up sitting on the sidelines and eventually they lose their jobs. The ones with rare talent and a lot of money invested in them will continue to get opportunities for a few years, but if they aren't a high pick or a guy with a lot of money invested in them they quickly disappear. In contrast, there are some guys who can play hurt and do so effectively. They may be injured but they are not injury prone because they continue to trot out there and do well.

 

[Tanner9919]

Just like the myth of the 'injury prone' player, even a cursory glance at the data totally debunks the idea that there's a 'rookie wall' for RBs.

Eight rookie RBs have had 250+ touches* since 2002 (when the FBG database starts including half-season splits):

Joseph Addai

Cadillac Williams

Kevin Jones

Willis McGahee

Domanick Davis

William Green

Clinton Portis

Domanic Rhodes

--All eight of these RBs got more touches in games 9+ than they did in games 1-8

--Average touches among the eight jumped from 14 in the 1st half to 23 in the 2nd half

--Seven of the eight RBs saw their points/touch rise during the 2nd half* of the season (Davis dropped from .64 to .62)

--Average FF points/touch for all eight backs rose 19% from the first half of the season to the second -- .58 to .69 (non-PPR)

The 'rookie wall' is a myth. Rookie RBs get stronger as the season goes on.

* Note that the FBG 2nd-half splits include playoff games.

not so fast my friend...your data is skewed..most,if not all rookie RB's play (very) sparingly in the first few weeks of the season..

Clinton Portis' rookie season ('02) saw him get 128 carries first 9 weeks of the season,avg of 14.2 per game.

weeks 10-17, he rushed 145 times, avg 20.7 per game.

Curtis Martin Avg'd 19.25 carries/gm over first 8 weeks of his rookie season ('95), and a whopping 26.75 over the final 8 weeks..

 

[CalBear]

not so fast my friend...

your data is skewed..most,if not all rookie RB's play (very) sparingly in the first few weeks of the season..

Clinton Portis' rookie season ('02) saw him get 128 carries first 9 weeks of the season,avg of 14.2 per game.

weeks 10-17, he rushed 145 times, avg 20.7 per game.

Curtis Martin Avg'd 19.25 carries/gm over first 8 weeks of his rookie season ('95), and a whopping 26.75 over the final 8 weeks..

...which is the opposite of what you'd expect if a rookie wall existed.

 

[PolishNorbi]

not so fast my friend...

your data is skewed..most,if not all rookie RB's play (very) sparingly in the first few weeks of the season..

Clinton Portis' rookie season ('02) saw him get 128 carries first 9 weeks of the season,avg of 14.2 per game.

weeks 10-17, he rushed 145 times, avg 20.7 per game.

Curtis Martin Avg'd 19.25 carries/gm over first 8 weeks of his rookie season ('95), and a whopping 26.75 over the final 8 weeks..

...which is the opposite of what you'd expect if a rookie wall existed.

Not really. Don't get me wrong I don't believe in the "rookie wall" but that doesn't prove anything. It proves that in the first half the of the season the rookie RBs are usually trying to win the spot. Once they win the spot, the coach isn't likely to give it that easily away. Here is quick proof:Giants, 2004:

Eli Manning ATT weeks 1-10: 9

Eli Manning ATT weeks 11-17: 189.

However, Manning put up HORRIBLE numbers for the first few games. 17/31 162 yrds, 6/21 148yrds, 12/25 113yrds, 4/18 27yrds. The coach didn't sit him after any of these games.

Edit *adding*: So if you a rookie wall existed you would have to check this way:

Find when they are near their peak of touches (finding if they got the full start), then split that part of the season in half or thirds or quarters than compare. However, this would be bad because it the sample size is way too small.