Letters to the Editor
-
Nice Joke
Nice joke at Carl Everett's expense. He's one of the great unsung assholes among modern athletes.
-
Floyd and Faith
Whenever an athlete tests positive, they always protest their innocence (the notable exception being the cyclist David Millar, who tested positive for EPO and fessed up) therefore such protestations are not useful as a discriminator of truth. In seeking to separate faith from reason, it may be useful to consider the fact that: 1) all Floyd's other tests during the tour were negative; 2) that testosterone is generally used over the course of training not as an isolated application (esp. when one knows testing will be performed), and most important; 3) that the French lab that performed the test ALWAYS leaks positive results to the press prematurely. Can we trust a lab that consistently behaves in an unethical manner and fails to follow its own policies. In contrast, Justin Gatlin's positive result was made public only after the appropriate confirmation and notifications. The difference between justice and a witch hunt is in how the hunt is carried out.
-
Leaked Test Results are a Good Thing
The only agenda the labs have in leaking test results is to prevent the rich and powerful who fail their tests from covering them up. I have no doubt that people who worked at these labs during past Olympics have plenty of stories about positive tests that the public never heard about. I'm sure Floyd's high priced team of lawyers had all sorts of plans and experts to testify about his supposed high natural testosterone level but once the info that his sample had synthetic testosterone in it his whole defence fell apart. If it hadn't been leaked no doubt it would have been ruled inadmissible in some court 6 months from now.
-
Stats and Testing
There's a very concrete statistical interpretation to "Faith and Testing." Take the following example:
Assume that the test is 95% accurate in all cases: so if an athlete is cheating, it catches the offense (tests positive) with 95% accuracy. Similarly, if the athlete is not cheating, it exonerates him (tests negative) 95% of the time.
Assume you test a random cyclist and you have prior beliefs that 3% of cyclists dope (this is the article of faith). Then a positive test is actually more likely a false positive--in fact, it's about 63% (if my arithmetic is correct) likely that you've falsely accused the cyclist. Similarly, if you have a much stronger belief that a cyclist dopes, then that 63% chance of falsely accusing someone drops.
From my very limited understanding of these tests, this is not at all unreasonable as these tests are simply not that accurate (by not that accurate, I mean 95 or 99 percent). Not like DNA testing where the accuracy of the test trumps anything but the most extreme prior beliefs about guilt.
So unless dope testing has made miraculous advances in accuracy in the last couple of years, King has it exactly right when he says it's a matter of faith that determines whether the party is guilty or not. Good article, great intuition.
-
Floyd Landis Never Takes His Hat Off...
I believe baldness usually is caused by high testosterone. For example, Greek men take great pride in their extraordinary manliness. (Women who love Greek men seem to agree.) And, nearly every Greek man I ever have met or seen is bald.
Floyd Landis never takes his hat off. He must be bald, even though he is quite young. When did he start losing his hair? All of a sudden, recently? Would that show that he recently started taking testosterone?
Are Floyd's brothers and cousins also prematurely bald? Would that show that he is just a natural stud, blessed by nature and genetics? Is Landis a Greek name?
-
Sick about the whole thing
The whole Landis thing makes sick. I don't really believe him now, but I haven't completely given up hope. I was one who celebrated his great stage victory in this space a few weeks ago, and I can't quite let myself give that up. I believed Pete Rose for far too long as well.
Part of the reason I believed Floyd at first is because the whole thing didn't make any sense to me. Why would he dope with something like testosterone on the day he was going to try to win a stage? He knew that he would be tested if he won the stage.
Plus, why testosterone? Until I looked into it a little more I thought that testosterone was only good for building muscle mass over the long term. Why then would he dope with it on race day? Apparently it can give an energy boost as well. So at least there is a logical reason for Floyd having done it.
But how could Landis have been so stupid to have done this when he had to know he likely be caught? If he did it, then I would guess it was about pride. He had to make up for his poor performance the previous day, no matter the cost. Now it looks like that cost will include his reputation and quite possibly his career.
I still want to believe Floyd, but unfortunately I can't. And that sucks.
-
wages of wins
Not sure about your joke about the Wages of Wins here (Rodman being better than Jordan.) But you don't have to read their book to know they don't argue that--check out their blog.
http://dberri.wordpress.com/2006/06/03/a-clarification-on-rodman-and-jordan/
Again, I haven't read the book and, in any case, the math would be way beyond me, but their point is that rebounding is undervalued and scoring--particularly inefficienct scoring--is overvalued.
Otherwise, good column as always. Yours is one of the very few columns of any type I read daily, along with one of the few sports anythings I read regularly.
-
Jon could use a math class
If a test is 95% accurate and a test comes back positive or negative, there is a 95% chance that the test is accurate; this remains true regardless of what percentage of testers are doping. This is basic math and logic. The reason this is true is because we are only talking about the validity of one test, whose accuracy will always match that of the accuracy of the test. A different outcome would result from the question: "if 100 samples are tested what is the chance that one will be erroneous."
In this case, you would also multiply the chance that the test is innacurate by itself given that there was the A & B tests. Thus if the test was 95% accurate, you would multiply .05 by .05 and have .0025 or 1 in 400. anyway, this is so stupid since none of this calculates the chance that the test was tampered with, which is Landis's only defense.
Anyway, King is totally right, who among us really has the slightest idea what's going on here.