Biopassport failure to flag Armstrong – advanced class

fer the advanced kids raggin on the entry level explanation of the ABP software

for the deeper understandin check WADA operating guidelines

and the free text review

then consider this figure

with actual ABP software output:

so

at the basic level

the blood values

are inputted into the software

and 2 things happen

1. the software uses the info

to adapt the thresholds

(the 99.9 confidence limmit)

in an effort to tighten the thresholds

based on the athletes physiology

2. the software the plots the new blood data

and flags values that fall outside the thresholds

the intent is that the software screens for abnormal profiles and flags them so that the profile can go on to expert review.

…

the calculation of thresholds

is called an adaptive model

because while it starts with population norms

(see how the high and low red threshold lines start very wide)

as additional data becomes available

the model automatically adapts

(note how the gap between the high and low narrows)

the method originally proposed

was based on Bayesian inferences

which would open the possibility

to add multiple factors that further improve

the predicted confidence limmits

for example

the population based threshold

ranges from 12 to 17

but

once a rider has a hgb measured

at

13

it becomes much less likely

that their natural value

would ever reach 17

therefore

the upper limmit can be lowered

(at some point it the bayesian approach was not fully implemented or watered down somehow but the details are not entirely clear from the operating guidelines)

ironically as can be seen in the outputs above

slow changes

such as might be seen with long term micro dosing

has the effect of shifting the riders thresholds

to adapt to their doped physiology

while tapering the dope

would have the opposite effect

ie

dope fast or come off fast

get popped

otherwise it’s steady as she goes
…

now

with regards to armstrongs data

this author don’t have the software

and as such

can’t exactly run armstrongs data through it

and

inserting this discussion

into

that post

would have lost the attention

of anybody with a life

so

compromises were made

in simplifying the adaptive thresholds

down to a best fit aproach of static values

so that the

concept

(that the software flags extreme high and low values

not patterns of elevated or suppressed values)

could be illustrated

to

a much broader audience

veloclinic is the reincarnation of one of our favorite Twitter friends, @captaintbag1, whose tumblr blog posts were a kind of blank verse, Tecate-soaked haiku of truthiness that cut through the slick bullshit and to the very core of what is gloriously fucked up about the sport of cycling. Although the Cap may be gone (sort of), his Doctor tbag/Captain Hyde alter ego lives on, and we’re glad to share his pithy analysis here. Lest you think these are the idiot ramblings of a madman, we’d like you to know that the doc is a legitimate professional in the science of sports medicine, and a savant when it comes to doping analysis. You have been warned.