What If It Wasn’t Floyd?

Dr. Arnie Baker said something over the weekend that has me thinking about who really tested positive at the Tour. What he said was this: One of the numbers in the lab documentation pack, which was similar to Floyd’s but not the same number, actually belonged to a rider who was tested after Stage 19. A different rider. And now, over at Steroid Nation, we can see a copy of a slide from a recent presentation by Dr. Baker (also displayed below) that shows that, yes, there was another rider whose sample number from Stage 19 (995475) differs only slightly from Landis’ sample number on Stage 17 (995474).

And further, Dr. Baker tells me the time period during which the samples from the two stages were tested overlaps. Leading to the very real possibility that some of the data from another rider’s tests could be mixed in with Landis’ data. In other words, it may not be so crystal clear whose test came up with the suspicious readings that LNDD claims shows the use of artificial testosterone.

Up to now many have assumed that the sample number 995475 shown in Landis’ lab documentation package was a typo and that it was really Floyd’s result. But this new twist suggests that this may not just be a typo, but rather data from another rider, from another stage that was incorrectly reported in Landis’ results. As you can see on Dr. Baker’s slide (shown below), Landis’ sample number on Stage 19 was labeled 994080. It now appears that because of some mistake, these results from Stage 19 may have gotten mixed up with Floyd’s data from Stage 17.

While it’s still possible that someone may have been doping with testosterone and tested positive at the 2006 Tour de France, this new revelation draws into question who that someone might be.

This shines a whole new light on the case. And it begs the question: Could it be that the biggest screw-up at LNDD didn’t have anything to do with laboratory technique, that it had to do with record keeping — as in: keeping track of who’s sample belonged to who?

If you go back and look at the so-called “whistleblower” documents, you’ll find that LNDD over the years has made a number of these kinds of mistakes. And in those instances, they’ve had to retract their conclusions about the tests performed. Although the authenticity of the letterhead the whistleblower documents were printed on has been drawn into question, no one — not even the head of LNDD — has denied the authenticity of the information, itself. In fact, what I’m hearing is that Dr. de Ceuarriz has grudgingly admitted that the information in the whistleblower documents is accurate. So it’s not too far-fetched to imagine that another bookkeeping error occurred in the Landis case. Will LNDD have to retract their findings for Landis, too?

From the time that the lab documentation pack became public, one question that I’ve had was whether or not riders had the same lab ID number throughout the race. That might make sense from an expediency point of view. But it didn’t make sense in terms of guaranteeing anonymity. After a while, if the same number kept turning up for analysis, someone could figure out who the rider was, and if he or she were so inclined, spike the sample. That’s a prospect that chills my soul.

So it makes more sense if the riders are given a new number with each sample they provide. That is a greater guarantee that the rider’s identity will be unknown to lab personnel. And from what we can see in this new slide, the number assigned to Landis’ sample after Stage 19 was not the same as from Stage 17. It was an entirely different number. So his anonymity could have been kept. But there was one other item that could give away who he is, and that was his therapeutic use exemption for using cortisone, due to his (formerly) bad hip.

How could the system give out two numbers that were so close? That gives me pause, as it’s a very predictable source of confusion in record keeping. All it takes is someone incorrectly recording for sample 995474 where data for 995475 should go, or vice versa, for data to become mingled together that shouldn’t be. And that’s what now appears to be likely.

If that did happen, it draws into question the integrity of all of the data. While that may sound like a getting off on a technicality, it’s a huge “technicality.” If you can’t be sure where the data came from — what tests, what person, what date — then you can’t draw any conclusions based on that data. Period. So as James at Steroid Nation says, the onus may now be falling on LNDD to prove that the data from the lab documentation pack attributed to sample number 995475 is, in fact, Landis’ data and not that of another rider.

Will that be a hurdle they can jump over? It’s hard to say. On the other hand, it may be yet another nail in the coffin for the anti-doping case against Floyd Landis. Time will tell.