jemima sumgong went from being a 2.29 marathon runner ,to 2.20.are these improvements normal for an EPO user?

No, I understood perfectly well what you meant. Your second go is no more logical than the first. You've merely substituted "basic understanding" for "fundamental notions." Such fuzzy notions are no substitute for actual math. Again, it's an attempt to make your posts sound rooted in science and math but there is no actual science of math underlying it, just rhetoric. Let me be more clear. The distribution of times can be exactly the same under the thesis "EPO works" and under the thesis "EPO doesn't work" as the existence of a statistical outlier has no explanatory value for the efficacy of EPO.

Then responding with "Expected distributions come from formulas and equations, not "fundamental notions."" makes no sense.I did not mean to say that expected distributions come from fundamental notions. I said that my ideas come from these fundamental notions.If "EPO works a lot", then I would expect "the distribution of times can be exactly the same" to be a false statement.

Mr. Obvious wrote:

No, I understood perfectly well what you meant. Your second go is no more logical than the first. You've merely substituted "basic understanding" for "fundamental notions." Such fuzzy notions are no substitute for actual math. Again, it's an attempt to make your posts sound rooted in science and math but there is no actual science of math underlying it, just rhetoric.

Let me be more clear. The distribution of times can be exactly the same under the thesis "EPO works" and under the thesis "EPO doesn't work" as the existence of a statistical outlier has no explanatory value for the efficacy of EPO.

That's where expected distributions come from. If you want to make some sort of point about EPO based on expected distributions, you should actually do the math and show your work. Just saying it is based on your ideas about fundamental notions is a nonsense statement that doesn't mean anything.You seem to have a very poor understanding of math and statistics. In particular you seem to have a poor understanding about what conclusions you might be able to show from one data point at the very tail of a distribution. This is not where the power of statistics lies.

You could right shift or left shift the distribution of times depending on how much you believe that EPO works. It would not alter the shape of the distribution or the existence of outliers.

rekrunner wrote:

If "EPO works a lot", then I would expect "the distribution of times can be exactly the same" to be a false statement.

passport era ,

do you work for iaaf / wada

as that reply sure sounds like cover to me

using words like... big fish caught...and ....fortunately for anti doping ......

wada the same people that encourage doping

and no where in that article nor you explained how CERA is just like

the other previous two versions of epo.

I am telling you now that CERA is alot better than other generations of epo

It was rocket fuel and the few selected athletes that had it handed to them by people within bid pharma proves this point .

CERA is magic and dynamite .

and enough with your rubbish attempt at disinformation

passport era wrote :

There's no differences between the first gen EPOs (alpha & beta) & CERA in terms of potency. The difference is the dose schedule of the two needed to achieve desired Hb/Hct levels. CERA has an added chemical to produce a much longer half-life. First gen EPOs require multi-day weekly injections vs once or twice monthly injections for CERA to achieve desired levels. The shorter interval of CERA administration can produce more RBCs and increase Hb levels. CERA was developed for people with disease pathologies (e.g., renal failure, anemia) that required an ESA but didn’t want the inconvenience of having to do injections multiple days per week (plus the more injections the risk of infection increases). Athletes don't enjoy injecting themselves very often either, so that's probably why CERA became a popular doping ESA with them when it came out in 2006/07.

The risk athletes have with CERA is it's long half-life of 140 hrs + vs 19 hrs (subcutaneous) & 7 hrs (IV) for first gen EPO. This is why when CERA became available and a subsequent test was developed with the help of the manufacturer, there was a rash of big name athletes from athletics to cycling testing positive. Fortunately for anti-doping, a top athlete such as Ramzi switched to CERA and was nailed after the Beijing (he had been avoiding a positive test with first gen EPO when he was registering those high off-scores of 148 & 157 during the 05 WCs, and WADA was routinely target testing him from that point onward).

Though they caught a big fish like Ramzi, consider how many athletes must have gotten away with doping by "not" using CERA.

But you asked me where my ideas come from, not where expected distributions come from.I answered where my ideas come from, and you respond with where expected distributions do and don't come from.Your responses indicate you are confused about both what you asked and what I answered. Despite you saying you understood perfectly, it seems you still don't.So tell me, why does everyone else get to make some sort of point about EPO, causing the one data point at the very tail of the distribution, without doing any math and showing their work? If someone says, EPO causes 9 minutes of benefit, or saying the real question is what kinds of drugs did Paula take to run 2:15, why isn't "Mr. Obvious" saying, hey wait a minute -- show us your math, show your work, show us your assumptions?Like everyone else, I can form an opinion about the potential causes, and unlikely causes, of the tail. I can expect that both clean peak performance levels, and EPO benefits, can be characterized by random distributions, without having to provide the equations. A 3 minute gap from these two identified sources would be an extremely low probability event, but of course, not impossible. I suppose, I could make assumptions, about averages, standard deviations, distributions, etc., and do some numerical calculations to illustrate the point. But this would be highly dependent on my assumptions. And it is not necessary, because thanks to my mathematical and statistical ability and intuition, I already know the outcome -- a low EPO benefit scenario is more likely to produce a big outlier, due to other, unique causes, independent of talent and drugs, than a high EPO benefit, that would mask or overshadow these other causes.If this is a nonsense statement that doesn't mean anything, then all opinions about EPO causing improvements in the marathon are nonsense statements that don't mean anything, because they all have the same basis, based on someone's notions of something where they do not have any data, and have not yet done the math and shown their work.

Mr. Obvious wrote:

That's where expected distributions come from.

If you want to make some sort of point about EPO based on expected distributions, you should actually do the math and show your work.

Just saying it is based on your ideas about fundamental notions is a nonsense statement that doesn't mean anything.

You seem to have a very poor understanding of math and statistics. In particular you seem to have a poor understanding about what conclusions you might be able to show from one data point at the very tail of a distribution. This is not where the power of statistics lies.

Not quite. You only get to right shift the points of known EPO users, by how much "EPO works".In that scenario, we can right shift Jeptoo, Shobukhova, and now Sumgong by some amount.Despite some strong opinions otherwise, we still don't know if we can right shift Paula at all. Assuming the worse, we can extend this exercise to include blood transfusions, and/or micro-dosing, allowing us to right shift different amounts.From the data we do know, Paula's highest blood values were not as extreme as Shobukhova's -- not even close.So in order for EPO to explain the extreme outlier we do have, we have to make a number of assumptions, especially in the "EPO can work by a lot" scenario.

Mr. Obvious wrote:

You could right shift or left shift the distribution of times depending on how much you believe that EPO works. It would not alter the shape of the distribution or the existence of outliers.

rekrunner wrote:

If "EPO works a lot", then I would expect "the distribution of times can be exactly the same" to be a false statement.

Rekrunner / Coach C - Why not just argue that:1. Some athletes are just outliers period. They are a cut above other runners. Place Bolt, Radcliffe, and El G/Bekele/Komen/ Haile in that group if you wish.2. That EPO works generally but response will vary on an individual basis and it's hard to predict the response because of a variety of factors. Many people may not agree with that but it's rational. And it's better than insisting on pseudo science and shifting arguments / explanations that frustrate people who just have a genuine interest in the sport and the prevalence of doping.I think Rekrunner got into all this debating because of the Paula thread.

"I'm making poor arguments without evidence or logic. You can't attack those arguments unless you attack all poor arguments without evidence or logic"

rekrunner wrote:

So tell me, why does everyone else get to make some sort of point about EPO, causing the one data point at the very tail of the distribution, without doing any math and showing their work? If someone says, EPO causes 9 minutes of benefit, or saying the real question is what kinds of drugs did Paula take to run 2:15, why isn't "Mr. Obvious" saying, hey wait a minute -- show us your math, show your work, show us your assumptions?

Random distributions? You have never take a stats class, have you?

random variables and student t distributions.

yyy wrote:

random variables and student t distributions.

I'm not sure if he is failing at English or math, but that does not seem to be what he is describing, but I will let him speak for himself.

3...2...1 wrote:

Rekrunner / Coach C - Why not just argue that:

1. Some athletes are just outliers period. They are a cut above other runners. Place Bolt, Radcliffe, and El G/Bekele/Komen/ Haile in that group if you wish.

2. That EPO works generally but response will vary on an individual basis and it's hard to predict the response because of a variety of factors.

Agreed - with the exception of Radcliffe, who is the only one of that list who could not keep up with the elite at age 20 - 25.

She is also the only one of that list with such crazy blood values + changes, as far is known.

Yeah her "natural talent" I laughed at that one, I'm glad you spotted the "EPO in preparation topped up with fresh blood before races." That is what I guessed from first seeing get the data. I do wonder whether some of her readings do smell alot like Xenon gas, but that is pure speculation.

There is a good chance that Mr Salazar was on something when he did 2:08.

aduck2002 wrote:

makoulfi was the only athlete that got done for micera

on condition kept his world titles , others too big .

keep repeating myself regardless of other peds this stiff is dynamite.

just look what it did for alan webb

and him before and after , just not even close to same athlete

and bekele an amazing athlete becomes unbeatable , run whenever

whereever any time of year any surface ,conditions .

again not quite same after 2008 when had to switch to normal epo

now on igf-1 lr3 also for marathon so cant say not chronic doper

and better but still not micera good

..................................................................................

...........................................................................................

There was something dodgy about his pulling out of the 800m injured, then invincible over 1500m, in house testing I suspect, plenty of water before testing?

Are you hinting that a certain little Ethiopian's fiancée mysteriously dropped dead from an overdose of CERA? Taking CERA injections like regular epo injections? Shocking, never, impossible. Nothing at all suspicious there. I would never believe it, just smears and innuendo.

It's a massive double standard. Compare the resistance and outrage and hostility and distrust and level of scientific standard demanded from Renato, who lacks credibility among his skeptics if he doesn't publish decades of blood data, against the ease of acceptance and unchallenged suggestions made in this thread that EPO caused Sumgong's 9 minute improvement.

Mr. Obvious wrote:

"I'm making poor arguments without evidence or logic. You can't attack those arguments unless you attack all poor arguments without evidence or logic"

rekrunner wrote:

So tell me, why does everyone else get to make some sort of point about EPO, causing the one data point at the very tail of the distribution, without doing any math and showing their work? If someone says, EPO causes 9 minutes of benefit, or saying the real question is what kinds of drugs did Paula take to run 2:15, why isn't "Mr. Obvious" saying, hey wait a minute -- show us your math, show your work, show us your assumptions?

I took a stats class at university. I got an A. Although I majored in Computer Science, I took a handful of advanced math courses, getting A's in all of them.What's wrong with "random distribution"?Oxford dictionary: "(Statistics) a probability distribution, especially the Poisson distribution."Wolfram Alpha: "A statistical distribution in which the variates occur with probabilities asymptotically matching their "true" underlying statistical distribution is said to be random."It was meant to describe any generic statistical distribution. Do you find it too specific? I saw no sense getting bogged down in the specifics and applicability of Normal, Gaussian, Poisson, Binomial, Cauchy, Weibull, "student t-"?, etc. distributions. Replace "random distribution" with "some statistical distribution", and the main point still stands. What kind of EPO efficacy distribution would produce a 3 minute single outlier for more than a decade, for a highly effective, widely prevalent, and difficult to detect drug? Do we see other examples, in other events or sports, of a single, drug induced, outlier?

Mr. Obvious wrote:

Random distributions? You have never take a stats class, have you?

1) Agreed -- no need to argue2) Maybe, maybe not -- the evidence to date that EPO works to improve performance in long distance running is not broad enough to encompass elites. The mechanism seems not well understood. Some say it's not so much about blood, but about recovery enabling harder training. Studies look at blood. Maybe it's both, up to a point. No one has attempted to produce a model of expected EPO benefits, versus any variable, beyond an average of study participants. It's easy to predict a diminishing return with higher talent, but hard to predict the behavior at the extreme top, due to lack of data.The Paula thread was a catalyst, but my opinion about Paula is actually incidental -- I'm not a fan of Paula as much as I am defending how to interpret the known facts within the light of the anti-doping science. That thread lasted so long because of all the backlash. I think it's a mistake over-emphasizing details like 24 degrees versus 30 degrees, calling it a lie for emotionally dramatic effect, talking about dehydration, and debating 30 minutes versus 90 minutes, when the science is clear, that hemo-concentration (not just dehydration, but also plasma shift out of circulation, into interstitial tissues), can occur at 23 degrees, within as little as six minutes, before you break a sweat, and takes two hours to return to the pre-exercise baseline. Taking blood within two hours, even when rehydrating, is known to produce inflated values, even after a 67 minute effort in 24 degrees. I think it's a mistake to completely ignore the science that shows the importance of machine brand differences, and machine calibration, before comparing two different blood values where such things are not known. The list goes on and on and on, as did the Paula thread. We should place more weight on statements backed up by peer-reviewed papers, and less on statements appearing in sensational newspapers and documentaries, where the backing is not plainly obvious, and not plainly unbiased.