Doping inspector backs Armstrong

[Mellow Velo]

An example of this is arguing that the lab which re-tested Armstrong's '99 samples tampered with the samples. Is there any evidence of this? Not that I've seen, or have seen discussed. About the only thing I've seen discussed is whether or not it is possible.

This is what I asked yesterday...with no results.

How many samples need to be EPO tampered with, to constitute an LA smear by the French?
6? No, of course not.
The answer is one, from say the Sestrieres stage, or an ITT.

So, job done and viola! I have a 6/8 to 1 chance of being successful.
Gambler's odds, but possible.

Attempting to contaminate 6 samples would give a roughly even chance of one hit, with LA and 5 other riders.

Getting 6 direct hits from 6 would be like winning the lottery.
Not the soundest of basis for a planned conspiracy.

 

[bianchigirl]

And that's without including the other 2 borderline samples - 8 samples spiked? All from one rider? Now, what are the odds?

 

[RTMcFadden]

And that's without including the other 2 borderline samples - 8 samples spiked? All from one rider? Now, what are the odds?

According to Ashenden

However, Lance Armstrong made that claim. Now, it's very easy to go back and assess the possibility of that scenario. We know the laboratory could not have known which samples belonged to Lance Armstrong. And we also know from the results, how many of Lance Armstrong's samples had EPO in them, and when during the race it occurred. Now the odds of the laboratory randomly selecting Lance Armstrong's samples out of those 87 samples, and let's just do it conservatively, just 6 times, 6 times they got his samples correct out of 87 possible tubes, the odds of that occurring are at least 1 in 300.

But I don't think that's correct. Anyway, that's for randomized sampling. If the blind is broken, then the odds are 1 in 1.

 

[progressor]

I've said my piece regarding some of the bias around here. I'll only acheive a banning from here if I keep responding to attacks.To me the playing field is clearly not even, and I can see I'm not the only one who recognises it that way - which is good.

If those who are having a fine old crack at me have any legitimate point it is getting of the topic of threads. I don't believe that's totally accurate with this thread, but it's not entirely unfair if you look across the range of threads I've become engaged in.

I'll do my best to constrain my arguments more directly to the threads topic, and consider the feedback forum for wider issues.

Moving on

According to Ashenden

However, Lance Armstrong made that claim. Now, it's very easy to go back and assess the possibility of that scenario. We know the laboratory could not have known which samples belonged to Lance Armstrong. And we also know from the results, how many of Lance Armstrong's samples had EPO in them, and when during the race it occurred. Now the odds of the laboratory randomly selecting Lance Armstrong's samples out of those 87 samples, and let's just do it conservatively, just 6 times, 6 times they got his samples correct out of 87 possible tubes, the odds of that occurring are at least 1 in 300.

But I don't think that's correct. Anyway, that's for randomized sampling. If the blind is broken, then the odds are 1 in 1.

Could you explain that maths more fully?

 

[RTMcFadden]

The variability allows us to determine when LA took rEPO - as in the studies you already posted the short window to get an positive from a sample.

While you are correct that a +/- is acceptable - this would be almost impossible to do through spiking a single sample.

Read the link I posted as it clearly articulates it better than I ever could (and because I am getting ready to go for a ride).

I think this is a source of confusion. According to Ashenden, you start with 20 ml of urine and filter to create the retentate. Then you take 20ul of rentate to perform the test. Nowhere does Ashenden identify exactly how much retenate is actually produced on average. Based on the comment regarding the adjustment for optimal value, one would expect that at least 40ul of retentate is produced, as the test would need to be re-run. Unfortunately, I can’t find any additional information to clarify this issue.

However, if we consider that 20ul is required to run the test, then one would expect that the sample size of the urine be set so that the vast majority of samples would produce at least 20ul. I would expect the retentate to be 3 times the required test sample size, and would like something closer to 5, given the fact that I may need to "adjust for optimal value".

As an aside, he doesn't mention how he would adjust the sample.

 

[poupou]

According to Ashenden

However, Lance Armstrong made that claim. Now, it's very easy to go back and assess the possibility of that scenario. We know the laboratory could not have known which samples belonged to Lance Armstrong. And we also know from the results, how many of Lance Armstrong's samples had EPO in them, and when during the race it occurred. Now the odds of the laboratory randomly selecting Lance Armstrong's samples out of those 87 samples, and let's just do it conservatively, just 6 times, 6 times they got his samples correct out of 87 possible tubes, the odds of that occurring are at least 1 in 300.

But I don't think that's correct. Anyway, that's for randomized sampling. If the blind is broken, then the odds are 1 in 1.

But the blind were not broken, at least there is nothing that proves it. It's a serious accusation that need a much more serious proof than those needed for a doping accusation isn't it?

 

[python]

the theory that the french spiked 1999 samples makes no plain common sense.

why would the french waste so much energy spiking 6 samples in 1999 when there was no epo test if they could easily accomplish armstrong’s destruction by a single epo spike in 2001-2005 and 2009 when the epo test was perfected.

fadden u sound like a crude obfuscator. i propose to introduce this word into english in your honor.

 

[Anonymous]

According to Ashenden

However, Lance Armstrong made that claim. Now, it's very easy to go back and assess the possibility of that scenario. We know the laboratory could not have known which samples belonged to Lance Armstrong. And we also know from the results, how many of Lance Armstrong's samples had EPO in them, and when during the race it occurred. Now the odds of the laboratory randomly selecting Lance Armstrong's samples out of those 87 samples, and let's just do it conservatively, just 6 times, 6 times they got his samples correct out of 87 possible tubes, the odds of that occurring are at least 1 in 300.

But I don't think that's correct. Anyway, that's for randomized sampling. If the blind is broken, then the odds are 1 in 1.

I have to agree with python, why didn't the same lab spike another sample later on instead of samples EVERYONE knew would not result in a doping infraction regardless of what they contained?

Secondly, please provide for us the proof that only Armstrong's samples were used out of those 87. In fact, there were 12 positives out of the samples used (again, please cite a source that reveals the total number of samples tested). And this is the BEST PART!!! See, the reason we know about Lance's samples and not the identity of the others is because LANCE PROVIDED HIS CONTROL NUMBERS!!! The journalists still don't know who the other positives came from because they nobody else was stupid enough to provide their numbers.

Now, being the big important scientist that you are, why would you make an assertion about the samples and not provide a source for your claims. Please cite it for us. Oh wait, maybe you didn't know that 12 samples were positive? 6 of those were Armstrong's and the identity of the other 6 are still unknown.

Dang, this learning curve thing...

 

[RTMcFadden]

But the blind were not broken, at least there is nothing that proves it. It's a serious accusation that need a much more serious proof than those needed for a doping accusation isn't it?

That's not true. The blind was broken by Damien Ressiot. The question is, was anyone else able to break the blind? I don't have proof either way.

 

[Anonymous]

That's not true. The blind was broken by Damien Ressiot. The question is, was anyone else able to break the blind? I don't have proof either way.

No, you don't. You just have an obfuscatory post meant to blow smoke up everyone's tush.

So now you are asserting that there was not only a conspiracy in the lab, but that someone with the UCI provided the control numbers prior to retesting so that someone could go to the trouble of spiking a sample having used gallons of water with EPO in it all reduced down to a drop so that they could produce EPO positives in 8 of Armstrong's samples (only they only got it right 6 times) that they knew would never result in a sanction because there were no B samples to test because these were the B samples and anyway they were just using the samples because they were working on perfecting the test anyway and then they had to be really sneaky and get a reporter to ask Lance for the numbers (which he provided) so that they could act like they didn't know already thereby throwing everyone off of their amazing plot to produce urine samples that contained synthetic EPO to thereby sully the reputation of an honest, humble man who just wants to ride his bike and cure cancer. That is your claim? Okay then...

 

[Anonymous]

And since we are posting things for which there is not proof or source, let me throw out one:
I think the 6 samples were spiked because Armstrong has tested positive on many occasions, and the people in the lab were getting angry because the UCI wouldn't do anything about it, so the UCI told them they could spike some of his old samples that couldn't be used for sanctioning Armstrong, but could sully his reputation a bit which would obviously satisfy the blood lust those lab workers have for Mr Armstrong which made everyone happy in the end because everyone got just a little piece of something that in the end had absolutely no consequences for anyone involved.

I for one wish there would have been some conspiracy to commit murder by one or the other parties so that John Berendt could have written a book about it. But still, since we are making unsubstantiated claims, that is mine...hey RTM, did I mention that I don't find anything you have written compelling?

 

[Anonymous]

Oh, by the way RTM, I don't find anything you have written to be compelling.

 

[RTMcFadden]

Could you explain that maths more fully?

This is not my area, but let me try. I used the excel function PERMUT to perform the permutation for the sample sizes. Also, I used College-Cram (http://www.college-cram.com/study/algebra/presentations/244) to work out the correct formula. The equation is as follows:

6 = Number of Armstrong Positive Samples
81 = Number of negative (non-Armstrong) Samples
87 = Number of total samples

6*(81!) / (87!) = 1.65 x 10 e11 = 165 trillion to 1

 

[Race Radio]

This is not my area

That is obvious.

The chance that there was this complex, French, Conspiracy is 0%. That Armstrong was doing what most of his top competitors were doing is 100%

 

[RTMcFadden]

No, you don't. You just have an obfuscatory post meant to blow smoke up everyone's tush.

So now you are asserting that there was not only a conspiracy in the lab, but that someone with the UCI provided the control numbers prior to retesting so that someone could go to the trouble of spiking a sample having used gallons of water with EPO in it all reduced down to a drop so that they could produce EPO positives in 8 of Armstrong's samples (only they only got it right 6 times) that they knew would never result in a sanction because there were no B samples to test because these were the B samples and anyway they were just using the samples because they were working on perfecting the test anyway and then they had to be really sneaky and get a reporter to ask Lance for the numbers (which he provided) so that they could act like they didn't know already thereby throwing everyone off of their amazing plot to produce urine samples that contained synthetic EPO to thereby sully the reputation of an honest, humble man who just wants to ride his bike and cure cancer. That is your claim? Okay then...

If by obfuscatory you mean I’m trying to pierce the veil that some hold that science is sacrosanct. Then yes I am. Scientists are people. People make mistakes. Therefore, Scientists make mistakes. Also, Scientists are fairly smart and are pretty good at figuring things out. Especially if they have 6 years or so to do it. In addition, I believe reasonable people can reasonably disagree.

I didn’t level the charge that there was a conspiracy, Armstrong did that himself. I’ve been trying to avoid casting dispersions regarding anyone’s motives. Because, fundamentally, I don’t really care if Armstrong doped or not.

 

[Mellow Velo]

Some further reading on this:-
http://nyvelocity.com/content/features/2009/spiking-armstrongs-99-samples
Lots of science and numbers....

 

[RTMcFadden]

I have to agree with python, why didn't the same lab spike another sample later on instead of samples EVERYONE knew would not result in a doping infraction regardless of what they contained?

Secondly, please provide for us the proof that only Armstrong's samples were used out of those 87. In fact, there were 12 positives out of the samples used (again, please cite a source that reveals the total number of samples tested). And this is the BEST PART!!! See, the reason we know about Lance's samples and not the identity of the others is because LANCE PROVIDED HIS CONTROL NUMBERS!!! The journalists still don't know who the other positives came from because they nobody else was stupid enough to provide their numbers.

Now, being the big important scientist that you are, why would you make an assertion about the samples and not provide a source for your claims. Please cite it for us. Oh wait, maybe you didn't know that 12 samples were positive? 6 of those were Armstrong's and the identity of the other 6 are still unknown.

Dang, this learning curve thing...

What I was trying to point out, is that the Lab was conducting research regarding the testing method(s). As part of that process, or more specifically, the method validation process, spiking samples as part of a recovery study is normal. I then suggested that these spiked results may have been inadvertently released. This way, there's no conspiracy, just honest people making an honest mistake.

 

[Bag_O_Wallet]

That is obvious.

The chance that there was this complex, French, Conspiracy is 0%. That Armstrong was doing what most of his top competitors were doing is 100%

If the samples were spiked, wouldn't the numbers be the same or near to the same across the board for all 6? Instead, tests seem to look like there's a higher concentration at key points in the race.

 

[RTMcFadden]

Some further reading on this:-
http://nyvelocity.com/content/features/2009/spiking-armstrongs-99-samples
Lots of science and numbers....

Thanks, I've actaully been focusing more on that. As you said, lots of science and numbers.

 

[Ferminal]

If the samples were spiked, wouldn't the numbers be the same or near to the same across the board for all 6? Instead, tests seem to look like there's a higher concentration at key points in the race.

I think the spikers made 8 different buckets of diluted EPO to make it look genuine.

 

[RTMcFadden]

If the samples were spiked, wouldn't the numbers be the same or near to the same across the board for all 6? Instead, tests seem to look like there's a higher concentration at key points in the race.

If the samples were spiked for a recovery study, the point would be to determine how much was detected per the method. In other word, I know how much I put in, so, how much did I get back. If I put in 100%, then I expect to see 100% back. That's not always the case. Sometimes you have interference from the sample matrix that prevents complete recovery. Sometimes you have variablilty within the test method technique. Gel electrophoresis methods are notoriously less robust than LC methods. Primarily, I believe, because it so manual / labor intensive (read as little automation). So, the variability of +/- 12.3% would make me a little uncomfortable, but would not necessarily invallidate the method.

 

[python]

What I was trying to point out, is that the Lab was conducting research regarding the testing method(s). As part of that process, or more specifically, the method validation process, spiking samples as part of a recovery study is normal. I then suggested that these spiked results may have been inadvertently released. This way, there's no conspiracy, just honest people making an honest mistake.

except again you sound like a crude obfuscator who's trying to get out of the pit he dug for himself. clearly u'r not familiar with the test processes.

when labs spike positive samples to evaluate unknown samples they don't do it to 86.666% isoforms. positive spiked samples are 100% positive and are certified by other labs. again your lack of knowledge proves that you're a crude obfuscator.

 

[RTMcFadden]

except again you sound like a crude obfuscator who's trying to get out of the pit he dug for himself. clearly u'r not familiar with the test processes.

when labs spike positive samples to evaluate unknown samples they don't do it to 86.666% isoforms. positive spiked samples are 100% positive and are certified by other labs. again your lack of knowledge proves that you're a crude obfuscator.

Unless, of course, you're the inovator. In that situation, you don't have any else to compare against.

 

[Digger]

Chances of samples being spiked:

First occasion Lance was positive, all four samples contained EPO, chances were four out of four. Second day, five samples, one positive, so chances were one in five. Third day, six samples tested, only one positive, so one out of six. Fourth day, four samples and two positives, so chances were two out of four. Fifth day, two out of four chance. Sixth day, two out of four.

Possibility of spiking LA's SIX positives:
4/4 x 1/5 x 1/6 x 2/4 x 1/4 x 2/4 = 16/7680 or 1 in 480

 

[Anonymous]

What I was trying to point out, is that the Lab was conducting research regarding the testing method(s). As part of that process, or more specifically, the method validation process, spiking samples as part of a recovery study is normal. I then suggested that these spiked results may have been inadvertently released. This way, there's no conspiracy, just honest people making an honest mistake.

Please cite the source showing that they were "spiking" those samples to validate the test.

Why wouldn't they have done that with samples they were sure didn't already contain EPO? Wouldn't that be the scientifically responsible thing to do? Wouldn't that protocol be the most logical one? I am thinking yes, it would. See, they weren't sure if there was EPO in the TdF samples, so why would they have used those in the manner you suggest. I am pretty sure, and you can check me on this, but if they just needed urine to spike, they could have gotten those samples from anywhere including their own urine which they would have known DIDN'T contain synthetic EPO.

When you get tired of being wrong, you can just stop posting.