Re: Dope Testing - An utter waste of time

On Jul 30, 12:29 pm, "b...@xxxxxxxxxxxxxxxxx" <b...@xxxxxxxxxxxxxxxxx>
wrote:
On Jul 30, 8:18 am, Davey Crockett <r...@xxxxxxxxxxxxxxx> wrote:





Idiots like Davey have recently been doing some arithmetic on the
probability of a rider who tested positive actually being a druggee or
alternatively whether the positive finding was in fact a "False
Positive" and various other deductions made based on same data.

However, these computations were based on the test proceedure having a
very high degree of accuracy. Even at 99 percent the chances that a
positive test result would actually "nail" a doping rider were to say
the least somewhat less than acceptable.

Admittedly the results we were discussing presupposed a "given"
percent of druggees in the universe to be tested - like we were using
an estimated 5 per cent of riders in the Tour de France "Universe"

But along comes a Cocky young Italian who says "Sure I charged up, but
your tests ain't worth *** because I was tested Ten times and came up
Positive only Twice."

Hmmm. Throw out the 98-99 percent effectiveness we were assigning to
the test and change that to 20 percent.

Then revamp the calculations and all but the simple minded will
believe that you have to be awfully unlucky to get caught.

Davey believes it's high time to abandon testing altogether and save
the exhorbitant costs of the flawed testing that is currently being
carried out.

"Anti-Doping" is an Industry in its own right these days. and its
Practitioners will of course Loudly Blow their Trumpets about how
effective they are and what a Stirling Job they are doing.

But Davey says "Bull***, assholes. Get your Snouts and Trotters out of
our Trough. And get back in the Welfare Line."

Sheesh. Ricco should get a medal for his exposure of the Testing Fraud
that the Bastards are Socking to Bikies.

When you made assumptions about "98-99 percent
effectiveness" you confused sensitivity and selectivity -
basically the rates of false negatives and false positives.
They don't have to be the same.  In fact they are often
anti-correlated.  If the test criterion is some number
above a threshold, and you raise the threshold, you
make it more likely to get a false negative, but less
likely to get a false positive.

For example, suppose 30% of the population is positive,
and you have a test with a 20% detection rate (80% false
negative) and 1% false positive rate.  If you test 1000
people, of whom 300 are positive, you'll find 60 true
positives and 7 false positives.  I made these numbers
up, but if we knew the real numbers we could get a
better idea.

Because false positives are undesirable, test thresholds
tend to be set kind of high, which leads to people not getting
popped on 100% of the tests they take.

What this really proves is that Ricco, taking dope regularly
enough that he feels he should have been busted all
ten times, was too cheap to pay for a good doctor who
would tell him how to microdose it and time the doses.
Forget whether he doesn't deserve to win the TdF because
he's Dirty Ricky the Cheating Doper.  He doesn't deserve
to win because he hasn't shown enough respect to
cheat well.

Ben- Hide quoted text -

- Show quoted text -

I thought it was a "tracer molecule" that was placed in the C.E.R.A by
the manufacturer that got The Cobra busted. If that is true, and the
molecule wasn't present, it may have been 10 out of 10 escapes and
he'd be on the podium.
--
Marty
.

Quantcast