Reason for Lemond's decline

[elapid]

As you indicated, I was not a coauthor of the paper in question. In fact, I had graduated and moved from Austin before Armstrong ever shifted his focus from triathlons to cycling, much less was tested in the UT-Austin lab. I also wasn't involved in the editorial process that led to publication of the paper. Thus, I can only comment as someone who 1) knows Ed Coyle, and 2) reads (and contributes to) the scientific literature (and thus knows the process).

Anyway, with respect to the questions you raise my take on the subject has always been if I had been a reviewer of the paper, I would have had to weigh the obvious flaws and limitations in the data against the novelty of/interest in the results. Given everything that has happened since, it is impossible for me to say which side of that equation I would have come down on back when the paper was first submitted - however, as I have expressed to various individuals (including Dave Martin, a coauthor with Ashenden on the letter-to-the-editor of JAP) I think the point is now really moot, i.e., the paper was published, various scientists responded, charges were leveled and dismissed, and when the dust settled the paper still wasn't required to be retracted by the journal editors. You can read into that what you may, but the bottom line is that if somebody really wants to refute the general finding (i.e., that efficiency improves over the course of a professional cyclist's career), the way to do that is with additional, contradictory data. Indeed, this is precisely what I told Dave Martin after he said that they had longitudinal data on a number of elite Australian cyclists. Interestingly, however, such data has never been published, whereas other studies have recently appeared which provide support for Coyle's results re. Armstrong....

Well, let me put it a different way. As a scientist, are you satisfied with the results and conclusions of this paper when two-thirds of the conclusions are obviously incorrect and manipulated (ie, weight loss and power improvement)? Again, I will leave the discussion regarding efficiency to more learned people than me.

I am on the editorial review board of three journals and an ad hoc reviewer for seven other journals in my field. I am not sure of the quality of the review and editorial process with the Journal of Applied Physiology, but to let such fundamental flaws go unanswered and uncorrected because of either novelty or interesting results is dubious at best. Particularly when the "interesting results" are so obviously incorrect (again, the claims of weight loss and 18% improvement in steady state power). I personally would be embarrassed if I was the author, reviewer or editor of this paper.

 

[Digger]

He certainly is eminent in his field, so if I were an attorney seeking an expert opinion to bolster my case I would be tempted to employ him as such.
QUOTE]

And with all due respect, I would be extremely reluctant to do the same in light of his inaccurate findings, and in one or two instances, downright fabrications.
In relation to the impact science has on the lives of paypeople, and their conversations, people are entitled to know the truth. Scientists should be at the forefront of this quest on so many levels. SHOULD being the operative word.
I honestly believe Coyle started off with the utmost intentions, I really do. But in light of his findings, it is apparent he was influenced by outside factors later in his study. It seems he was told to find favourable results by others. And in fairness he wouldn't be the first person this has happened to, in light of the person I am referring to.

 

[unsheath]

Why should I know what happened to the data when I had graduated and moved away from Austin long before Armstrong ever set foot in the lab?

Are you saying this was never discussed in person with Coyle? I find that far fetched given how central the data is to the findings. Have you not questioned him yourself? Do you not have a duty of care to do so?

 

[beroepsrenner]

I'm no academic but for what its worth I can relate to my own experience in regard to this debate. Although I have no recorded data, the earlier part of my racing career was spent concentrating on track racing. After transforming myself into a roadie, my body fat percentage dropped considerably with a corresponding increase in sustained power output and a dramatic improvement in climbing ability. This occurred well before experimenting with PEDs.

 

[Kennf1]

I'm no academic but for what its worth I can relate to my own experience in regard to this debate. Although I have no recorded data, the earlier part of my racing career was spent concentrating on track racing. After transforming myself into a roadie, my body fat percentage dropped considerably with a corresponding increase in sustained power output and a dramatic improvement in climbing ability. This occurred well before experimenting with PEDs.

So you're saying that your training focused on road racing, and as a result, you got better at road racing?

 

[acoggan]

Are you saying this was never discussed in person with Coyle? I find that far fetched given how central the data is to the findings. Have you not questioned him yourself? Do you not have a duty of care to do so?

No, I have not discussed this particular fact with Ed...we don't really communicate all that often (a one or two sentence email a couple of times a year?), and the topic simply hasn't come up.

As for any "duty of care", I don't see why you believe that I should be appointed the world's policeman...

 

[acoggan]

Well, let me put it a different way. As a scientist, are you satisfied with the results and conclusions of this paper when two-thirds of the conclusions are obviously incorrect and manipulated (ie, weight loss and power improvement)? Again, I will leave the discussion regarding efficiency to more learned people than me.

As I stated previously, I'm not sure precisly how I would have rated the paper had I been one of the original reviewers.

I am on the editorial review board of three journals and an ad hoc reviewer for seven other journals in my field. I am not sure of the quality of the review and editorial process with the Journal of Applied Physiology

It is widely considered the most prestigous journal of its kind, with the editorial board historically reading like a "who's who" of exercise/environmental/respiratory physiology.

, but to let such fundamental flaws go unanswered and uncorrected because of either novelty or interesting results is dubious at best. Particularly when the "interesting results" are so obviously incorrect (again, the claims of weight loss and 18% improvement in steady state power). I personally would be embarrassed if I was the author, reviewer or editor of this paper.

So in essence you are claiming that you know more about how to evaluate the quality of the data and the paper than do those in the field.

 

[TRDean]

The Journal of Applied Physiology only has an impact factor of around 3.632. If that is the best in the field...the field needs some improvement.

 

[acoggan]

I honestly believe Coyle started off with the utmost intentions, I really do. But in light of his findings, it is apparent he was influenced by outside factors later in his study. It seems he was told to find favourable results by others.

I think that you are wrong. After all, the conclusions of the study are logically consistent with a number of other studies out of Ed's lab re. cycling efficiency and the role of fiber type, and the paper was published well before the SCA lawsuit came to trial. IOW, if you don't think he was fully objective in his assessment and/or reporting of the data, you should be looking elsewhere for an explanation.

 

[acoggan]

The Journal of Applied Physiology only has an impact factor of around 3.632. If that is the best in the field...the field needs some improvement.

Comparing impact factors across fields is not valid.

 

[TRDean]

Comparing impact factors across fields is not valid.

Thats Bull Shiiiit!!! The Journal of Applied Physiology is not a great journal. Period!!!!! I do know this! It is a good well thought of journal.

 

[acoggan]

Thats Bull Shiiiit!!! The Journal of Applied Physiology is not a great journal. Period!!!!! I do know this! It is a good well thought of journal.

A quote from one of many assessments of the impact factor:

"Extension of the impact factor to cross-discipline journal comparison is... inappropriate."

http://www.ncbi.nlm.nih.gov/pubmed/14668520

 

[TRDean]

A quote from one of many assessments of the impact factor:

"Extension of the impact factor to cross-discipline journal comparison is... inappropriate."

http://www.ncbi.nlm.nih.gov/pubmed/14668520

who is comparing across disciplines...you? I never said that...did I? NO!!

 

[elapid]

So in essence you are claiming that you know more about how to evaluate the quality of the data and the paper than do those in the field.

Not in essence - there is no quality to this data. I do not appreciate people trying to claim superiority because of their profession or how long they have been a PhD. So you can step off your high horse and address the issues raised rather than trying to avoid my specific questions because I am not some world renowned physiologist. I am a scientist and world renowned in my own field. I know how to evaluate data. I have readily admitted that I do not know about efficiency and its calculations and that as a result I would not comment on this. But this was not my question to you. For the questions I posed in this particular case, it wouldn't matter if you were a plumber or a Nobel laureate, the raw data is there and anyone with a modicum of common sense and simple mathematical skills can evaluate the data in this paper.

So back to my original questions which you did not answer about how the raw data provided in the paper and table 2 was misused. So please answer the questions rather than be a politician:
- How can Coyle claim a reduction in body fat was responsible for Armstrong's increased power-to-weight ratio when he didn't measure body fat and lean body weight was actually higher than preseasons 1992 and 1993?
- How can Coyle use an estimated and unmeasured racing weight to calculate a power-to-weight ratio from a preseason power in 1999, but not in 1992 and 1993? This deliberate error resulted in the false 18% improvement in power-to-weight ratio when this improvement is actually 6.9% compared to 1992 and 1.6% compared to 1993 preseasons.

These are pertinent questions because Coyle claims "Therefore, over the 7-yr period, an improvement in muscular efficiency and reduced body fat contributed equally to a remarkable 18% improvement in his steady-state power per kilogram body weight when cycling at a given Vo(2) (e.g., 5 l/min)." Two out of these three conclusions can be thrown out just by reading his paper and doing some basic calculations.

 

[acoggan]

who is comparing across disciplines...you? I never said that...did I? NO!!

You did precisely that in your first post, in which you stated that if a journal with an impact factor of "only" 3.X is the best of its field, then that field needs improvement.

 

[acoggan]

Not in essence - there is no quality to this data. I do not appreciate people trying to claim superiority because of their profession or how long they have been a PhD. So you can step off your high horse and address the issues raised rather than trying to avoid my specific questions because I am not some world renowned physiologist. I am a scientist and world renowned in my own field. I know how to evaluate data. I have readily admitted that I do not know about efficiency and its calculations and that as a result I would not comment on this. But this was not my question to you. For the questions I posed in this particular case, it wouldn't matter if you were a plumber or a Nobel laureate, the raw data is there and anyone with a modicum of common sense and simple mathematical skills can evaluate the data in this paper.

So back to my original questions which you did not answer about how the raw data provided in the paper and table 2 was misused. So please answer the questions rather than be a politician:
- How can Coyle claim a reduction in body fat was responsible for Armstrong's increased power-to-weight ratio when he didn't measure body fat and lean body weight was actually higher than preseasons 1992 and 1993?
- How can Coyle use an estimated and unmeasured racing weight to calculate a power-to-weight ratio from a preseason power in 1999, but not in 1992 and 1993? This deliberate error resulted in the false 18% improvement in power-to-weight ratio when this improvement is actually 6.9% compared to 1992 and 1.6% compared to 1993 preseasons.

These are pertinent questions because Coyle claims "Therefore, over the 7-yr period, an improvement in muscular efficiency and reduced body fat contributed equally to a remarkable 18% improvement in his steady-state power per kilogram body weight when cycling at a given Vo(2) (e.g., 5 l/min)." Two out of these three conclusions can be thrown out just by reading his paper and doing some basic calculations.

So if your opinion is such that anyone with a "modicum of common sense" would have to come to the same conclusions, then what you are saying is that the editors of JAP must lack such common sense since they not only accepted the paper in the first place, they did not insist on its withdrawal after the results were questioned. IOW, what you see as fatal flaws in the paper experts in the field in question clearly did not. Why do you suppose that is?

(As for me, I realize that the scientific process, like any human endeavor, is imperfect, and don't see anything to be gained by getting my shorts all in a bunch over this particular publication and the process that led to it.)

 

[BroDeal]

So if your opinion is such that anyone with a "modicum of common sense" would have to come to the same conclusions, then what you are saying is that the editors of JAP must lack such common sense since they not only accepted the paper in the first place, they did not insist on its withdrawal after the results were questioned. IOW, what you see as fatal flaws in the paper experts in the field in question clearly did not. Why do you suppose that is?

(As for me, I realize that the scientific process, like any human endeavor, is imperfect, and don't see anything to be gained by getting my shorts all in a bunch over this particular publication and the process that led to it.)

LOL. This is classic. Steadfastly refusing to address any of elapid's points, making the argument that the paper must be good because it was accepted by JAP, and finishing up with the reasoning that even if the paper crap, the errors should be ignored because it still might advance science.

It reminds me of the see no evil, hear no evil, speak no evil approach you used to take about doping on Usenet.

 

[elapid]

So if your opinion is such that anyone with a "modicum of common sense" would have to come to the same conclusions, then what you are saying is that the editors of JAP must lack such common sense since they not only accepted the paper in the first place, they did not insist on its withdrawal after the results were questioned. IOW, what you see as fatal flaws in the paper experts in the field in question clearly did not. Why do you suppose that is?

(As for me, I realize that the scientific process, like any human endeavor, is imperfect, and don't see anything to be gained by getting my shorts all in a bunch over this particular publication and the process that led to it.)

You are entirely missing my point and yet again have not answered my questions.

The letters to the editor regarding this paper were all regarding the efficiency calculations, not the two points I have raised. I said in my first post to you that I did not know why more had not been made of these two factors. I don't know why, and hence why I asked you. But I read the paper and look at the data and it is very obvious to me that Coyle made some glaring mistakes. In previous threads a few months back, other posters also had the same reaction when I posted the data. You can infer what you will, but rather than defending the integrity of the JAP and its editors, why don't you answer my questions?

I have asked you three times now, and three times you have not answered. Why do you suppose that is?

I'll post them yet again:
- How can Coyle claim a reduction in body fat was responsible for Armstrong's increased power-to-weight ratio when he didn't measure body fat in 1999 and lean body weight in 1999 was actually higher than preseasons 1992 and 1993?
- How can Coyle use an estimated and unmeasured racing weight to calculate a power-to-weight ratio from a preseason power in 1999, but not in 1992 and 1993? This deliberate error resulted in the false 18% improvement in power-to-weight ratio when this improvement is actually 6.9% compared to 1992 and 1.6% compared to 1993 preseasons.

Look at these questions. They do come down to common sense (body weight claims and inappropriate use of unmeasured racing weight) and basic mathematics (power-to-weight calculations).

 

[acoggan]

You are entirely missing my point and yet again have not answered my questions.

The letters to the editor regarding this paper were all regarding the efficiency calculations, not the two points I have raised. I said in my first post to you that I did not know why more had not been made of these two factors. I don't know why, and hence why I asked you.

And I thought that I answered you, i.e., clearly the reason is because the reviewers and/or editors did not consider these to be critical flaws.

 

[acoggan]

LOL. This is classic. Steadfastly refusing to address any of elapid's points, making the argument that the paper must be good because it was accepted by JAP, and finishing up with the reasoning that even if the paper crap, the errors should be ignored because it still might advance science.

You seem to be confusing what I think happened with what I think should have happened.

It reminds me of the see no evil, hear no evil, speak no evil approach you used to take about doping on Usenet.

I just don't really see the point of speculating about things of which I have absolutely zero direct knowledge, especially when it contradicts my personal belief that every person is innocent until proven guilty. That is why I rarely enter into discussions of doping unless 1) I know something that could contribute to the discussion (e.g., how a GC/C/IRMS works and is used to try to differentiate between endogenous and exogenous testosterone, as in the Landis case) or 2) someone asks for my opinion (e.g., the recent article by Laura Weislo here on CN).

 

[elapid]

And I thought that I answered you, i.e., clearly the reason is because the reviewers and/or editors did not consider these to be critical flaws.

Are you a politician or a physiologist? I am asking YOU these questions, not what you think of the reasonings of the reviewers and/or editors.

As you are not Coyle and were not an author on this paper, I will try and ask these questions differently:

Do you think that these are errors in Coyle's paper?

If no to one or both points, specifically answer the following:
- If you think that the first point is not an error, please answer for me how YOU think it is acceptable that Coyle can conclude that Lance's improved power-to-weight ratio was because of reduced body fat when this was not measured or documented and his lean and gross body weights say the opposite?
- If you think that the second point is not an error, please answer for me how YOU think it is acceptable to use an unmeasured and estimated racing weight to calculate a power-to-weight ratio from a preseason power and then compare this power-to-weight ratios that were calculated using only measured preseason weights and powers?

 

[elapid]

And I thought that I answered you, i.e., clearly the reason is because the reviewers and/or editors did not consider these to be critical flaws.

This is not to detract from my previous post, which raises the most important questions I have long being trying to get you to answer, but I really don't know how much more critical you can get in regards to flaws when Coyle's paper concludes:

Improved Efficiency + Reduced Body Fat = 18% Improved Power-to-Weight Ratio

Out of this conclusion, improved efficiency has been questioned, reduced body fat is incorrect, and an 18% improved power-to-weight ratio is incorrect. These are some CRITICAL flaws.

 

[acoggan]

Do you think that these are errors in Coyle's paper?

Errors in the data or in its interpretation?

please answer for me how YOU think it is acceptable that Coyle can conclude that Lance's improved power-to-weight ratio was because of reduced body fat when this was not measured or documented and his lean and gross body weights say the opposite?

I'd say that the sentence you keep quoting is worded poorly, as clearly it is an assumption that Armstrong's self-reported loss of body mass was entirely due to a reduction in body fat. But worthy of the sorts of historonics exhibited by you and others here? Not in my book.

(BTW, since no data were collected in-season during Armstrong's Tour-winning years, you clearly cannot claim as you have that the data that are presented contradict the above assumption.)

- If you think that the second point is not an error, please answer for me how YOU think it is acceptable to use an unmeasured and estimated racing weight to calculate a power-to-weight ratio from a preseason power and then compare this power-to-weight ratios that were calculated using only measured preseason weights and powers?

To quote the original article (with emphasis added):

"Anthropometry. Total body weight during laboratory testing
ranged from ~76 to 80 kg from 1992 through 1997 as well as
during the preseason in 1999. However, when competing in the
Tour de France in 1999–2004, body weight was reported by
the subject to be ~72–74 kg. Lean body weight was ~70 kg
during the period of 1992–1997 (Table 2)."

IOW, Coyle was entirely up-front about he arrived at his numbers, so the question devolves to whether this limitation of the study is so significant as to preclude its publication. Obviously the reviewers didn't think so, or else the paper wouldn't have been published. As I've said a number of times before, I don't know precisely how I would have come down on the issue of accept/reject, and at this point it is impossible for me to say. I do think, though, that the standard for rescinding a paper is different than the standard for accepting one in the first place.

EDIT: I believe that one reason for my differing perspective is that, unlike so many others, I do not view the paper in question as either evidence for or against Armstrong being clean. Thus, the most interesting part of the study isn't what it says (or claims to say) about changes in his power:mass, but what it says about improvements in his efficiency. The Aussies (Ashenden in particular) don't particular care for that "take home" message, but 1) they have never addressed the changes in gross efficiency (which is what really matters from a performance perspective...delta efficiency is only relevant when it comes to discussing potential physiological mechanisms) and 2) a number of subsquent studies (not case reports) have provided data that supports Coyle's position.

 

[acoggan]

This is not to detract from my previous post, which raises the most important questions I have long being trying to get you to answer, but I really don't know how much more critical you can get in regards to flaws when Coyle's paper concludes:

Improved Efficiency + Reduced Body Fat = 18% Improved Power-to-Weight Ratio

Out of this conclusion, improved efficiency has been questioned, reduced body fat is incorrect, and an 18% improved power-to-weight ratio is incorrect. These are some CRITICAL flaws.

You statements above aren't really correct:

1. no one has really been able to bring into question the reported changes in gross efficiency (which is what really matters from a performance perspective). The Aussies tried, by claiming that the ergometer that Coyle said he used never existed, but they just made themselves look foolish in the process (since it clearly did, e.g., I used it in my dissertation).

2. since Armstrong's body composition was never measured when he said he weighed 72-74 kg, we don't know what his percent body fat was at that time. That is irrelevant, however, since power:mass is calculated from total mass, i.e., you don't need to know percentage body fat to calculate it. The fact that his in-season mass was based on self-report rather than direct measurements clearly is a limitation, but since A) Coyle made that clear right up front and B) the focus of the paper was really on changes in efficiency, not weight loss (e.g., read the Discussion...there isn't a single paragraph devoted to changes in body mass/composition, whereas four consider the changes in efficiency and the possible mechanisms), I don't think you can really consider it a critical flaw. It only seems that way to people (like, e.g., Ashenden) who think that the paper has anything to do with whether Armstrong doped or not.

 

[elapid]

Errors in the data or in its interpretation?

Interpretation. The data is there for everyone to see and is the reason why I bring up these two points. Without the data, we would not know that Coyle did not measure Lance's body fat percentage, nor how he came up with the 18% improvement in power-to-weight ratio.

I'd say that the sentence you keep quoting is worded poorly, as clearly it is an assumption that Armstrong's self-reported loss of body mass was entirely due to a reduction in body fat. But worthy of the sorts of historonics exhibited by you and others here? Not in my book.

Which sentence is it that I keep quoting which is worded poorly?

Armstrong's self-reported loss of body mass is not in question, nor is any assumption that this was because a reduction in body fat. What is in question is Coyle using "reduced body fat" as one of two reasons to explain Armstrong's 18% improvement in power-to-weight ratio when reduced body fat has not been measured.

(BTW, since no data were collected in-season during Armstrong's Tour-winning years, you clearly cannot claim as you have that the data that are presented contradict the above assumption.)

To quote the original article (with emphasis added):

"Anthropometry. Total body weight during laboratory testing
ranged from ~76 to 80 kg from 1992 through 1997 as well as
during the preseason in 1999. However, when competing in the
Tour de France in 1999–2004, body weight was reported by
the subject to be ~72–74 kg. Lean body weight was ~70 kg
during the period of 1992–1997 (Table 2)."

No data was collected during any in-season, not pre-1996 and not 1999-2005, so your point is mute. Coyle presents three preseason data sets, so their comparison is valid. What is not valid is Coyle using an in-season estimation for a preseason calculation and then comparing this to other preseason calculations.

Now, back to my questions. Are you going to answer them or are you going to argue semantics and keep squirming?