The New 18.5 Minute Gap

Or: How Much Longer Can This Go On?

At the end of March, I wrote about how, without access to the original data files, LNDD’s findings from Stage 17 — that Floyd Landis had tested positive for synthetic testosterone — could not be verified.

In what seems to be a never-ending series of lab mixups, LNDD appears to be the gang that can’t archive data straight. Team Landis sent out a press release earlier this evening which states that the original data files from Landis’ Stage 17 tests have been deleted from LNDD’s computers, thus destroying evidence that could potentially exonerate Landis. The evidence could also have proved fatal to the Landis defense, assuming it was authentic and the results were consistent upon reanalysis. But now, it appears that the authenticity of the data is in question. And with it, the Stage 17 A- and B-sample results.

Apparently, LNDD copied the data to CD-ROM. The problem is that the data bears the wrong timestamps. According to the Landis press release:

The data concerning the Stage 17 “A” samples were re-saved on 1/30/2007. Landis’ “B” sample data bore a time stamp of 4/26/07, 9:51 a.m. CET, prior to the scheduled arrival of the independent expert and Davis later that day.

I’m not sure how LNDD can now verify that the data is the original data, uncorrupted or unmodified. And, according to Team Landis, some of the data files are missing. If this is true, it may shoot a big hole in USADA’s case. With no way to independently verify LNDD’s results, it’s hard to believe that the case can stand up to the kind of challenges Landis’ legal team are certain to mount against USADA and the French lab.

But before we, the defenders of the Landis faith, jump for joy, there’s one thing we should remember: This isn’t a court case we’re talking about. It’s arbitration. And the rules in arbitration aren’t exactly the same as the rules in a court of law. So while a judge might throw out the case because the data can’t be proven to be original, the two arbitrators (McLaren and Brunet) who allowed the new B-sample tests might still allow this data as evidence against Landis. How would they justify doing so? Well, I can’t say for sure. But then, I didn’t think they’d allow the additional B-sample tests, and they found a pretty novel way to justify that.

What I’d like to know is more information about the other data which Team Landis tells us is missing from the files copied to the disk. Would this data be strong enough to impeach the test results? Hard to say without knowing what kind of data it is, or what it might represent.

Just as importantly, LNDD and USADA are now forced into a position where they have to defend the accuracy and integrity of their data. Whether they can do so to the “comfortable satisfaction” of the arbitrators is an open question. But the only people LNDD has to blame for being in this situation is themselves. If they’d properly archived everything so that the date and timestamps were unaffected, Team Landis would not be able to make an issue of the data’s accuracy.

TBV makes a good point, when, in a comment to his post Deleted/Tampered files at LNDD, he says:

The safe move, unless there is stuff we don’t know about, would be for Landis to treat [the data files] as correct, and do the analysis necessary to show/not show the problems they suspect. They still need to prove something. If they fail to find what they want, they can still claim tampering; but if they find what they want, then it doesn’t matter.

This latest revelation just adds to the impression of sloppy procedures at LNDD. Whether it will be enough to bring an end to the case, or exonerate Landis, is something we won’t know for at least a few more weeks.