FanGraphs Baseball


RSS feed for comments on this post.

  1. “According to the story, the test “won strong consensus among doping scientists and experts from around the world who attended a London symposium.” Again, a group which largely contains doctors and scientists who would appear to have something to gain from the test gaining a foothold in a league like the NFL (or MLB).”

    That doesn’t mean they’re wrong.

    Comment by MrKnowNothing — January 10, 2013 @ 4:36 pm

  2. The problem with tests like this is that the false positive rate depends not only on the rate at which the tests produce false positives, but the rate at which HGH is used, a quantity we can’t possibly know. For example, if no one is using, then all positives are false, if everyone is using, all positives are true.

    Still it’s good to see that they’re at least trying, I just hope no one’s reputation gets ruined for no reason.

    Comment by Stathead — January 10, 2013 @ 4:50 pm

  3. What is up with this article? If you have specific information that HGH testing is not reliable, you should present that, rather than questioning the motivation of the people involved.

    “Virtually indistinguishable” is not the same as “indistinguishable.” The fact that the current test is based on isoform differences between the naturally occurring and recombinant HGH shows that they are in fact distinguishable. Since the new test is based on a biomarker – testing for an effect of HGH – the main problem would probably be determining the threshold between a normal amount of HGH and an amount which indicates PED use. That doesn’t seem insurmountable, and I believe this is similar to how other doping tests work.

    Comment by BrianH — January 10, 2013 @ 5:05 pm

  4. I want the game to be as clean as possible. So I am willing to live with false positives. Players make millions of dollars a year. If they were to be given a false positive and suspended, they would still live a very comfortable life. The likelihood of one player getting multiple false positives in one career is tiny and worth the risk. Lets make MLB the cleanest game in the world! I think nearly all Americans would be proud if this were true!

    Comment by NATS Fan — January 10, 2013 @ 5:20 pm

  5. The thing is, the mainstream media continues to bring out quasi-experts to discuss matters like this, when they should go straight to true experts like college professors, like the Baseball Economist did, see this post at his Sabernomics website:

    That link and other links to studies finding no helpful effects from HGH are at this website, just search for the term HGH:

    And I would note, as far as I know, having “denser bones, thicker skin, less fat, and more lean body tissue” are not attributes that lead to better baseball performance. There are nearly 300 pound guys who are lollipop hitters while skinny sticks (like Aaron was when he was young) blasts homeruns all over the place. Thicker skin only helps with interviews :^). Less fat, well, we know of plenty of fat players who have done very well in baseball, Pablo Sandoval being the most recent rotund hero, Gwynn, Kirby Puckett, Boog Powell, Harmon Killebrew, heck, Babe Ruth. And more lean body tissue, well, that never helped Johnny “Boo” LeMaster much.

    And really, if there really were health benefits, punishing people because they take a medicine to heal themselves sooner? How is that different from a bandaid? Alcohol is just as addicting as many illegal drugs, but it is legal while the rest is illegal.

    The media is as much to blame for the steroids/PED era as any stakehholder in baseball, more so in my mind because only they among the stakeholders are suppose to be protecting the sport. Can’t expect the owners to do it, particularly since the Players Union has hamstrung any attempts to hold their players accountable for anything (really? allow a player back after 7 known abuses of cocaine? And they minimize any attempts at penalties), and the players are accountable for their own actions but given that it was economically incentivized to push them to use, I don’t exactly blame them, that’s human nature. And the fans are fans, we will watch, through thick and thin.

    No, it is the journalists who should have been watching the barn door. And the door has been wide open since at least the 60’s when Jim Bouton’s book came out about amphetamine usage, and nothing was done until the steroids hit the fan, and nobody did anything when McGwire was caught red-handed with Creatine in 1998. Nobody thought of investigative reporting back then? And the rumors were swirling even before that.

    This holier than thou attitude that pervades the national baseball writer scene is just sickening to me, they should have been protecting the interests of the sport and fans long ago, now is too little, too late, too moralistic, too sanctimonious, and really, too self-serving. They were the ones who were suppose to be investigating this stuff, they were the ones asleep at the wheel, and now they want to stand up and be counted? Hypocrites.

    Comment by obsessivegiantscompulsive — January 10, 2013 @ 5:27 pm

  6. Well that is true, but presumably controlled studies have been conducted to determine those rates.

    Comment by Macek — January 10, 2013 @ 5:30 pm

  7. Well that’s why you might test it on people where you do know who is taking it.

    Comment by KrunchyGoodness — January 10, 2013 @ 5:32 pm

  8. How do false positives create “the cleanest game in the world?” As you readily admit, false positives will cost the players millions of dollars, not just in salary but also in endorsements. It will also cost them publicly. And for what? This will simply create the image of being the cleanest game in the world. Players will still be ahead of the testing and others will be falsely exonerated. Let’s face it….if the tests are inaccurate enough to provide false positives, there will certainly also be false negatives.

    That’s nowhere near being “the cleanest game in the world.” That’s simply putting on a show for the public to fool them into thinking it’s the cleanest game in the world. Clearly, it’s already working.

    Comment by chuckb — January 10, 2013 @ 5:36 pm

  9. This was meant to be a reply to NATSfan above. Sometimes I love my Kindle Fire, and sometimes I just don’t.

    Comment by chuckb — January 10, 2013 @ 5:44 pm

  10. This reads like tinfoil hat conspiracy theory nonsense. Who would rather ask than the experts?! I’m sure plumbers would be pure of motive….

    Comment by Jason H. — January 10, 2013 @ 6:10 pm

  11. No. You don’t understand. You can get the sensitivity and specificity rates for the test and that’s all well and good, but you can’t know the probability of a false positive unless you know the prevalence of what you’re looking for in the population, i.e. the number of players using HGH. It’s an consequence of Bayes Theorem.

    Comment by Stathead — January 10, 2013 @ 6:28 pm

  12. True experts like college professors. Good one.

    Comment by maguro — January 10, 2013 @ 6:33 pm

  13. Spoken like a true Nazi.

    Comment by Baltar — January 10, 2013 @ 6:40 pm

  14. If I were king of the world, the world would be a much better place. And one way in which it would be better is that athletes would be allowed to do anything they want to their own bodies for our entertainment.

    Comment by Baltar — January 10, 2013 @ 6:44 pm

  15. Actually, creatine was and still is a legal supplement. But Mcgwire was “caught” with a bottle of Androstene, which was at the time a legal over the counter supplement, now considered a psuedo-steriod (because it “acts like a steroid” according to the FDA) and was pulled from shelves in 2004. Therefore, I can’t consider him being “caught” with anything, when at the time it was perfectly legal according to real law and MLB Policy, which at the time was very vague.

    Comment by Dave in GB — January 10, 2013 @ 7:03 pm

  16. I understand what you are saying from an entertainment perspective, but if this were actually the case, I would never want to be a professional athlete, or for my children to be professional athletes (or athletes at all). These things might be great for short term performance, but they are not good for you.

    Sports are a great way to encourage healthy exercise. Making PED an acceptable, expected, or necessary part of the culture robs this benefit.

    Comment by Jason H. — January 10, 2013 @ 7:04 pm

  17. Other than that, I agree with everything else

    Comment by Dave in GB — January 10, 2013 @ 7:06 pm

  18. Stathead,

    In a controlled experiment you do know the prevalence of HGH use in the population.

    Comment by Jason H. — January 10, 2013 @ 7:10 pm

  19. While there are certainly many professors that know nothing more about their subject than anyone else, there are still many professors who actually do credible and useful work. Like most areas of thought, balance is far more desirable than taking an extreme position such as the one you implicitly hold.

    Comment by Brad — January 10, 2013 @ 7:12 pm

  20. Very well said, ogc. Wow! You’re on fire with this one!

    Comment by DrBGiantsfan — January 10, 2013 @ 7:21 pm

  21. This is a PR stunt. If MLB were serious about drug testing they would adopt the current (since early 2000) WADA approach to drug testing, which has moved away from relying on direct testing (it doesn’t work), and now focuses on the indirect, long-term surveillance of individual athlete profiles. Instead of trying to find illegal substances, this approach collects urine/blood samples over time and monitors each individual athlete for “unnatural” fluctuations.

    So basically, the article is right: this is a PR stunt, because the world’s leading drug testing body isn’t really investing seriously in direct testing anymore. This is much more about the appearance of control, or appearing to do something, than actually giving a crap about finding drug cheats.

    Comment by Bryan — January 10, 2013 @ 7:36 pm

  22. Do I trust HGH testing? Howtheheck is anyone who is not a clinical chemist with additional training in endocrinology supposed to answer that question? It’s like asking Rush Limbaugh and Rachel Maddow to debate whether global warming is real!

    Comment by DrBGiantsfan — January 10, 2013 @ 7:45 pm

  23. Actually, this post sounds holier than thou–blame the sportwriter??. Remember what happened to Canseco when he wrote his book? He became a laughingstock. I remember thinking probably there was a lot of truth in what he wrote but then being surprised when it garnered nothing but derision. I mean, really, he just made all of it up?
    The PED debacle is squarely on the players with owners being major accessories. I do not see it as a sportswriters job to police the sport. It would have been better if some writer had the cajones to do it, but short of a smoking gun, the players would have circled the wagons, denied it to the end, and left the writer hung out as a fool. Someone like Gammons maybe could have pulled it off. Did he see injections with another impeccable witness, could he scoop up the syringes, and had them analyzed? The evidence would have had to have been absolutely airtight for any writer to do it and he would know he was risking his livelihood, one that doesnt pay millions. In fact there is so much $$ in MLB it is not farfetched to imagine players being able to buy witnesses (see Bonds, Barry). No, the players took the PEDs and the clean ones should have said Heck NO, but they did not.

    Comment by Pennant — January 10, 2013 @ 7:48 pm

  24. NATSfan–that post reads like satire.

    Comment by Pennant — January 10, 2013 @ 7:50 pm

  25. How, exactly, do doctors gain form this?

    Comment by snack man — January 10, 2013 @ 8:04 pm

  26. “scientific uncertainty is impossible to completely attain.” No, I attain scientific uncertainty on a daily basis.

    Requiring certainty would force use to remove the home run from play and study each out for if the glove really held onto the ball until the end of the Earth. Get over it, sometimes you just have to go ahead and play the game according to the rules and hope for the best.

    Comment by snack man — January 10, 2013 @ 8:12 pm

  27. The biomarker test is not what was agreed to by the MLBPA in 2011 as it predated WADA’s approval. The isoform test was the only approved WADA test at the time. Weiner specifically mentioned a 48-72 hr detection limit although in fact it is really more like 24 hrs with proper dosing which is why so few athletes have been caught using HGH.

    The biomarker test does not actually measure HGH, it measures HGH supposed effects. Such indirect tests could introduce more false positives.

    Comment by pft — January 10, 2013 @ 8:24 pm

  28. Mcguire has since admitted to using steroids and HGH.

    Comment by pft — January 10, 2013 @ 8:27 pm

  29. WADA’s new approach seems sound. Individuals have wide variations in hormone levels and requires a normal range to be rather broad.

    The Babe likely had a testosterone level that was off the charts at the time but was his normal, and men in the early 20th century had higher natural testosterone levels than today, so normal in a population changes over time. There are certainly individuals today who may be at the edge of the populations normal or outside. If a person is low normal, theoretically, he may cheat and still pass the screening test as the results are under the threshold limit set at some level above high normal values to account for uncertainty and minimize false positives.

    The testing should be geared to determining if someone introduced synthetic drugs to boost an individuals normal levels, and not to punish those whose levels are outside of the populations normal levels, or make it easy to cheat for those who have low normal levels.

    Unfortunately, diet and external influences may affect hormone levels and cause them to fluctuate. The science really is not that well understood, so I would wonder someone gets scrutinized for fluctuations not related to PED’s

    Comment by pft — January 10, 2013 @ 8:46 pm

  30. Thanks for the info about MLB/WADA testing negotiations.

    Of course biological profiling (indirect testing) doesn’t measure HGH, that’s the point: direct testing of HGH is a waste of resources. It really only serves a symbolic purpose. The fact MLB is resistant to long-term monitoring of indirect signs – the current gold standard in Olympic drug testing – speaks for itself.

    Direct testing is basically useless (from a pragmatic, cost-benefit approach); and indirect testing is scientifically questionable, as you point out. My angle on all this is that we need to have a realistic conversation about drug testing: what it can actually accomplish, what resources it would take to reach that goal, and potential costs (e.g., false-positive; privacy).

    Comment by Bryan — January 10, 2013 @ 8:53 pm

  31. Jason H. Yes. You do. That gives you some information, like if someone is using, the probability of the test diagnosing correctly, or if someone is not using, the probability of the test diagnosing correctly.

    But the MLB is not a controlled environment. If you want to know P(Using given positive test result) (in probability, “given” is written as | so that would be written as P(Using|positive test result) which I will abbreviate to P(U|+)) then you apply bayes’s theorem and you get P(U|+)=P(+|U)*P(U)/P(+).

    P(+|U), i.e. the probability of a positive test given using is determined from those controlled studies. P(+) you can also figure out by conditioning on using:
    P(+)=P(+|U)*P(U)+P(+|not using)P(not using).

    P(Not using)=1-P(U) so what you’re left with is this obnoxious P(U) which is the probability that the player is using, which can be roughly interpreted as the proportion of players that are using. If you take into account that HGH doesn’t actually give players benefits(? I’m learning today from the author and from commenters. On a side note, I’m okay with catching guys TRYING to cheat, even if they’re doing it wrong ala corked bats), then this probability is almost pure speculation, or at least will take a number of positive tests to determine. If the positive rates are equal to what you’d expect if no one is using, then there’s a good chance no one is using.

    I do this for a living I (hope I) know what I’m talking about.

    Comment by Stathead — January 10, 2013 @ 10:12 pm

  32. What position did I implicitly hold, other than that credentialism is a stupid way to approach an issue like this?

    Comment by maguro — January 10, 2013 @ 10:45 pm

  33. Stathead,

    Right, MLB is not a controlled environment. That is why you develop the test in a controlled environment where you can measure type I and II errors directly. ….you then assume the real world false positive, false negative and success rates are accurately estimated by the control, since you can’t know these directly (as you point out). …this is pretty reasonable and easy to do since there is a huge population of people that do and do not take HGH for medical reasons.

    You are over thinking this.

    Comment by Jason H — January 10, 2013 @ 11:23 pm

  34. and if you read my post, you’ll see I address that, but it only helps with part of the equation. The rate of use in MLB is going to be different than that in the general population.

    Comment by Stathead — January 10, 2013 @ 11:50 pm

  35. Stathead,

    The rate of use in the general population is irrelevant.

    You apply your HGH test to 1000 individuals you know are taking HGH and 1000 individuals you know are not taking HGH.

    Your test results:

    HGH + samples:

    999 +
    1 –

    HGH – samples:

    999 –
    1 +

    False positive = 1/1000
    False negative = 1/1000
    true positive = 999/1000
    true negative = 999/1000

    ….extrapolate empirical rates to unknown (i.e. mlb).

    Comment by Jason H — January 11, 2013 @ 12:32 am

  36. You very clearly don’t understand, and I’m not going to keep arguing with you. If no one in MLB is using HGH then any positive you get is a false positive. If everyone is using, then every negative you get is a false negative. In those cases it doesn’t matter how accurate your test is.

    Comment by Stathead — January 11, 2013 @ 3:28 am

  37. it’s good that MLB is testing in-season now but let’s be honest: those hgh-tests are useless, Lance Armstrong and hundreds of other cyclists, hundreds of other athletes in all sports proved that 10-15 years ago.

    Comment by Bavarian Yankee — January 11, 2013 @ 7:29 am

  38. Hey Stathead, am I understanding you correctly? I think that you’re saying that if the rate of use in the population are testing doesn’t approximate the rate of use in the population in which the test was validated you will have different false positive and false negative rates. To me that makes sense.
    My question is does that actually change the probability that one individual who should test positive does because to me it seems logically like it shouldn’t.

    Comment by Macek — January 11, 2013 @ 9:04 am

  39. Stathead,

    What you are saying is correct. Its just not an actual worry. Do you seriously believe that the test developers are unaware of the fact that they need to produce a test with a false positive rate so low that they can be confident that their positives are positive? They can know the false positive rate empirically. Even if no one uses HGH and all positives are false positives, if only 1 in 1000000 tests produces a false positive, you never produce a false positive because you don’t perform enough tests to where you are expected to do so.

    You are arguing that good drug testing is impossible. But because we can know the empirical error rate, and we can perform independent tests on B samples for all positives, and we know that athletes do, in fact, use drugs, we can test for it reliably.

    Comment by Jason H — January 11, 2013 @ 9:25 am

  40. No, it doesn’t. It might rob you, or your children, of the chance to compete as a professional athlete free of drug-use-pressure, but that’s all.

    You can still go shoot hoops with friends, get a great workout in, and not have to worry about PED usage – unless you’re playing some real fierce pickup games.

    Comment by MrKnowNothing — January 11, 2013 @ 10:42 am

  41. More work, money, exposure, legitimacy, authority. So, that’s actually quite a lot to gain.

    Comment by Neil S — January 11, 2013 @ 10:42 am

  42. Isn’t that why you run a second test? (or at least collect multiple samples). Run the test on one sample, positive.. instead of leaking it to the press, you run the test on the second sample. If that sample is negative, then the overall process is a negative result. If it too is positive, then we have a cheater ! That’s how you get around the false positive.

    And, if you wish, you can go collect the second sample at same time, or a short bit later (like at the earliest possible convenient time for both parties. Player wants a clear name, MLB wants a good test.. both should be agreeable)

    Comment by Cidron — January 11, 2013 @ 11:16 am

  43. Its not the false positives.. its the overall accuracy. If it is (grabbing a number) 97% accurate, that’s okay. but, it gives 3 in a 100 a quick “it was a false positive, i didn’t do HGH”. That is what they want reduced. They don’t want to look like fools any more than you or I do. And, they don’t want the process cluttered up with that, or appeals based on a “high error rate”. So, they want it as accurate as possible, with very little false positives. Multiple test samples, and other means/methods are of course involved to reduce this. We, and they want a clean game. But, you have to do it right, or the overall process is flawed, then it becomes abused then worthless.

    Comment by Cidron — January 11, 2013 @ 11:21 am

  44. Baseball writers are policing the sport. Who do you think feeds us the information that shapes our opinions? Who do you think the information leaks go thru? AND, who do you think controls the votes for the baseball hof? The writers do, fully, or at least heavily in all of those. They are controlling the sport as well as any other union, player, umpire, owner, management aspect does. They are the ones that shape our perception.

    Comment by Cidron — January 11, 2013 @ 11:26 am

  45. or, it could introduce more “direct” testing. If player A exhibits normal signs of .. hgh, then no additional testing is warranted. If he starts to exhibit (in your words, false positive) signs of hgh use (no accusation of it yet, just “he has ‘symptom x’ over time”) then a team is sent to collect a few samples for testing.. If that too fails, we have a cheater. Barry Bonds.. larger hat size (easily), send a team to collect a sample. Babe Ruth has a pot belly.. no reason.. he always had that.

    Comment by Cidron — January 11, 2013 @ 11:32 am

  46. and, the sciences of testing and catching them and uncovering cover-ups hasn’t advanced in the last 10-15 yrs …. right? What with all the other advances, this is the one that didn’t.

    Comment by Cidron — January 11, 2013 @ 11:33 am

  47. If there is a financial incentive to deliver an accurate test, it will be developed. Lab testing is a business, just like baseball. As a former Medical Technologist with years of experience, no test is 100% accurate.
    There can always be errors, so you can’t just dismiss any testing that may develope a false positive/negative. The goal is obviously to make that outcome as low as possible.

    Comment by Hurtlockertwo — January 11, 2013 @ 11:43 am

  48. And it’s a bit misleading too. It’s basically charging these scientists and experts with conflict of interest. Absolutely, some of these doctors and experts might be working for the company who does the drug tests (you would hope that they would attend these things to increase their knowledge and expertise) but it’s a bit much to level that charge at every member of the consensus.

    Comment by siggian — January 11, 2013 @ 12:04 pm

  49. Gee, why don’t you go all out and declare that everyone charged with a crime is automatically guilty. Sure, some people who didn’t actually commit a crime will end up in jail, but…

    Comment by siggian — January 11, 2013 @ 12:12 pm

  50. heck, nothing is 100% true/accurate/proof in the medical community.. Even the basic condom is only 99%… and of my four kids, two beat the condom (and one beat the pill). That percentage exists for a reason, other than a disclaimer vs lawsuits.

    Comment by Cidron — January 11, 2013 @ 12:59 pm

Leave a comment

Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Close this window.

0.265 Powered by WordPress