EPO reportedly delivers an enormous increase in physical performance and is very difficult to test for. Getty Images

A member of the US Olympic Committee whistled yesterday when he read a new report from Denmark about the staggering power of Erythropoietin, a.k.a. EPO.

You might say, he was whistling past the graveyard.

EPO gets less publicity than human growth hormone, but it's everywhere, ghost-like in its invisibility. Marion Jones used it for years without flunking a drug test, as did Olympic gold medal sprinter Andrew Pettigrew. Four-time All-Pro lineman Dana Stubblefield got away with injecting it until he finally copped to a charge of lying to the feds in January. And a trainer for the boxer Sugar Shane Mosley said the boxer used EPO as rocket fuel. Colombian cyclist Luis Herrera once famously remarked, "When I saw riders with fat asses climbing cols like airplanes [at the Tour de France], I understood what was happening."

But we apparently haven't realized how powerful it really is.

According to a study published yesterday in the Journal of Applied Physiology, a team of scientists from the Copenhagen Muscle Research Center gave EPO to eight young men to see its effect. What they found was that it increased performance up to 16 percent when the subjects were giving their maximum effort, and—get this—up to 50 percent at a slightly lower level of exertion. That's what caused the Olympic official to whistle.

"At this level, a percent or two of extra effort can win you a gold medal. Fifty percent? My God," he said. "We've never seen research like this before."

Now for the graveyard part.

With the Olympic trials in track & field and swimming starting tonight, and the Tour de France launching next week, we're going to be hearing a lot about how drug testers are closing the gap on the cheaters. The UCI, which runs cycling, has tripled its anti-doping budget to $8 million and promised to do 10,000 out-of-competition tests this season. But collecting the samples is only half the process. The other half is getting them analyzed. And that's where the Copenhagen group stumbled onto its real bombshell.

They sent 106 EPO-laced urine samples to a pair of labs accredited by the World Anti-Doping Agency. The samples were taken at three different times—during the 14-day "boosting phase" when EPO use was highest; the "maintenance period" when the subjects were tapering down; and in the post-treatment phase. Here's the scorecard:

Boosting Phase: Lab A found all the samples positive. Lab B found none.

Maintenance Phase: Lab A only found six examples of "misuse" in 16 samples—a 37 percent success rate. Lab B missed the boat again, classifying just three as "suspicious."

Post-Treatment Phase: Lab A found 2 out of 24 positive. Lab B found—you guessed it—none.

Anthony Butch, director of the UCLA Olympic Analysis lab in Los Angeles, says he wasn't surprised at the low numbers considering that his lab has done 2,600 tests and found only six positives. "EPO has a short half-life in the body and everybody knows what it is," he says. " That's why it's very hard to catch athletes taking it."

What surprised him was the gap between the two labs. At the point when the chances of catching the user was highest—in the boosting phase—Lab B completely whiffed. If I'm a doper, I want to be in the same country as that one.

Which raises the question. Where exactly is Lab B?

Who knows? The study's lead author, Carsten Lundby, fails to mention that in his report. And, when I asked him in an email, he refused to tell me, just like he's refused to tell WADA. "I think it's best that the labs remain anonymous," he wrote me, raising questions about exactly he hopes to achieve with the study.

But that's just one of the questions I wanted answered. Another is how, exactly, did he get the labs to do the tests in the first place?

Butch's Los Angeles lab only accepts samples from WADA or its accredited sporting federations. Lundby seemed positively glib when he told me, "I just asked if they would analyze my urine samples according to WADA regulations, and they answered, 'Yes.'"

Which begs another, even thornier question: If a WADA lab is willing to take samples from outside sources, can any athlete get his levels privately checked as a way of evading the very same drug tests that the lab will be analyzing for an anti-doping agency?

Lundby tried to calm those fears by saying that one of the two labs called the Denmark Anti-Doping Agency to make sure he wasn't doping professional athletes. He left me to guess about the other.

Howard Jacobs, a Los Angeles attorney who has represented a rogues' gallery of convicted dopers, from Tim Montgomery to Floyd Landis and Tyler Hamilton, doubts the study will change much on his side of things. "We already knew this was a flawed test," he says. "All this study tells us is that guilty athletes are slipping through the cracks, not that innocent ones are being falsely accused."

That's cold comfort. Just when it seemed these WADA guys had it all figured out, that they'd built a machine with a $26 million budget, along comes a study like this. As we enter the high season for drug testing, the questions this study doesn't answer are more worrisome than the ones that it does.

Shaun Assael is a Senior Writer for ESPN The Magazine