These were nobody's main points. No one in history has ever argued doping prevalence is as low as 1-2%. Their main point was to estimate the "true" doping prevalence.
Too many words and numbers for his decrepit brain.
It does seem that he has completely missed the main point, always acting as if the only possible doping discussion is his "main point" -- a point so trivial that no one discusses it.
You have just summoned rekrunner, as we see. He has been trying to discredit the survey - without success - for years here.
The main point, which is irrefutable, is that the incidence of doping at championship levels far exceeds the numbers caught.
Doping will certainly have continued to develop in the last decade, as much of it is based on advances in medicines and the rewards for doping remain a powerful incentive.
These were nobody's main points. No one in history has ever argued doping prevalence is as low as 1-2%. Their main point was to estimate the "true" doping prevalence.
My "success" is highly dependent on the reader. Independent of my success, the study speaks for itself, for those prepared to read all of it, and able to understand what the mathematical models represent. Their main point was to propose an anonymous survey as a useful tool to estimating "true" doping prevalence, while simultaneously identifying, but not resolving, no less than 8 barriers that would need to be overcome, standing between their survey results and true doping prevalence.
Much of the "debate" here over the years was about properly interpreting the study, whether the unrefined "43.6%", or post-processed "31.4%" after eliminating the fastest responses, was their "true" estimate, and whether the researchers rationalized that it was "likely more than" 43.6% or 31.4%.
But the rationalized opinions of the authors are not particularly relevant. Given all the unresolved unknowns identified in the study, there is no guarantee that any of the study results or opinions are "true" or even "likely". As we have seen, despite the often discussed 43.6% and 31.4% figures, the study itself presented no less than 39 potential estimates, in an attempt to eliminate potential distortions, ranging from 8.9% to 73.1%. Even these are not strict lower and upper bounds, but rather a result of an arbitrarily chosen 30% non-compliance. We have no way of identifying which of these 39 estimates are the closest, when determining the "true" prevalence requires resolving all of the unknowns, and passing the unfiltered results through a combination of all 8 mathematical filters.
You do realise this is exactly the same post you have made on this subject for years on end? You persuade no one. But your arguing the figures is irrelevant. They remain estimates, not a definitive finding. Nothing you say changes that. The key point remains that the actual incidence of doping - whatever estimate you take - is far higher than testing indicates, which shows that most dopers are getting away with it. That means doping is out of control. So the sport is as dirty as hell.
This post was edited 41 seconds after it was posted.
These were nobody's main points. No one in history has ever argued doping prevalence is as low as 1-2%. Their main point was to estimate the "true" doping prevalence.
My "success" is highly dependent on the reader. Independent of my success, the study speaks for itself, for those prepared to read all of it, and able to understand what the mathematical models represent. Their main point was to propose an anonymous survey as a useful tool to estimating "true" doping prevalence, while simultaneously identifying, but not resolving, no less than 8 barriers that would need to be overcome, standing between their survey results and true doping prevalence.
Much of the "debate" here over the years was about properly interpreting the study, whether the unrefined "43.6%", or post-processed "31.4%" after eliminating the fastest responses, was their "true" estimate, and whether the researchers rationalized that it was "likely more than" 43.6% or 31.4%.
But the rationalized opinions of the authors are not particularly relevant. Given all the unresolved unknowns identified in the study, there is no guarantee that any of the study results or opinions are "true" or even "likely". As we have seen, despite the often discussed 43.6% and 31.4% figures, the study itself presented no less than 39 potential estimates, in an attempt to eliminate potential distortions, ranging from 8.9% to 73.1%. Even these are not strict lower and upper bounds, but rather a result of an arbitrarily chosen 30% non-compliance. We have no way of identifying which of these 39 estimates are the closest, when determining the "true" prevalence requires resolving all of the unknowns, and passing the unfiltered results through a combination of all 8 mathematical filters.
Too many words and numbers for his decrepit brain.
The more waffle there is the more easily you are fooled. But you're only here as rekrunner's lap dancer.
Too many words and numbers for his decrepit brain.
It does seem that he has completely missed the main point, always acting as if the only possible doping discussion is his "main point" -- a point so trivial that no one discusses it.
You would never get the essential point, since your purpose is to argue what is false. You also mistake sprinkling verbal and numerical confetti for cogent argument.
You do realise this is exactly the same post you have made on this subject for years on end? You persuade no one. But your arguing the figures is irrelevant. They remain estimates, not a definitive finding. Nothing you say changes that. The key point remains that the actual incidence of doping - whatever estimate you take - is far higher than testing indicates, which shows that most dopers are getting away with it. That means doping is out of control. So the sport is as dirty as hell.
Your "key point" remains when the true doping prevalence is as low as 10%, or even 5%. If that is the only key point, as you insist, one has to wonder why do a survey at all, just to hammer the point that no one in history ever doubted. One also has to wonder, why does anyone argue with me at all about the true prevalence, when we all agree it is surely more than 2% found in testing?
Speaking of "key points", the researchers explicitly wrote, in a boxed section entitled "Key Points" that "we found that the prevalence of past-year doping was at least 30% at WCA".
But the key point of this thread is this figure: "43.6%". Whether you find me persuasive or not doesn't change any of the discussion and limitations expressed by the researchers themselves, and all the possible ways they suggested to remove any and all "distortions" built into that 43.6%, as a function of degree of survey non-compliance, to finally arrive at an accurate estimate of the "true" doping prevelance.
And ultimately having an accurate tool to measure doping prevalence will be necessary to help understand when doping gets under control.
You do realise this is exactly the same post you have made on this subject for years on end? You persuade no one. But your arguing the figures is irrelevant. They remain estimates, not a definitive finding. Nothing you say changes that. The key point remains that the actual incidence of doping - whatever estimate you take - is far higher than testing indicates, which shows that most dopers are getting away with it. That means doping is out of control. So the sport is as dirty as hell.
Your "key point" remains when the true doping prevalence is as low as 10%, or even 5%. If that is the only key point, as you insist, one has to wonder why do a survey at all, just to hammer the point that no one in history ever doubted. One also has to wonder, why does anyone argue with me at all about the true prevalence, when we all agree it is surely more than 2% found in testing?
Speaking of "key points", the researchers explicitly wrote, in a boxed section entitled "Key Points" that "we found that the prevalence of past-year doping was at least 30% at WCA".
But the key point of this thread is this figure: "43.6%". Whether you find me persuasive or not doesn't change any of the discussion and limitations expressed by the researchers themselves, and all the possible ways they suggested to remove any and all "distortions" built into that 43.6%, as a function of degree of survey non-compliance, to finally arrive at an accurate estimate of the "true" doping prevelance.
And ultimately having an accurate tool to measure doping prevalence will be necessary to help understand when doping gets under control.
As hard as it is for you to accept, analysis of the survey figures suggest the extent of doping at championship level in athletic sports could be more than 1 in 2 athletes. The lower figures could be 1 in 3. Either way, that's a corrupted sport for which antidoping has no effective answer since testing is missing most of those athletes.
These were nobody's main points. No one in history has ever argued doping prevalence is as low as 1-2%. Their main point was to estimate the "true" doping prevalence.
My "success" is highly dependent on the reader. Independent of my success, the study speaks for itself, for those prepared to read all of it, and able to understand what the mathematical models represent. Their main point was to propose an anonymous survey as a useful tool to estimating "true" doping prevalence, while simultaneously identifying, but not resolving, no less than 8 barriers that would need to be overcome, standing between their survey results and true doping prevalence.
Much of the "debate" here over the years was about properly interpreting the study, whether the unrefined "43.6%", or post-processed "31.4%" after eliminating the fastest responses, was their "true" estimate, and whether the researchers rationalized that it was "likely more than" 43.6% or 31.4%.
But the rationalized opinions of the authors are not particularly relevant. Given all the unresolved unknowns identified in the study, there is no guarantee that any of the study results or opinions are "true" or even "likely". As we have seen, despite the often discussed 43.6% and 31.4% figures, the study itself presented no less than 39 potential estimates, in an attempt to eliminate potential distortions, ranging from 8.9% to 73.1%. Even these are not strict lower and upper bounds, but rather a result of an arbitrarily chosen 30% non-compliance. We have no way of identifying which of these 39 estimates are the closest, when determining the "true" prevalence requires resolving all of the unknowns, and passing the unfiltered results through a combination of all 8 mathematical filters.
You persuade no one.
Whereas your 40,000 posts have persuaded everyone that you are a sad, angry, drunk and lonely old man without a life.
As hard as it is for you to accept, analysis of the survey figures suggest the extent of doping at championship level in athletic sports could be more than 1 in 2 athletes. The lower figures could be 1 in 3. Either way, that's a corrupted sport for which antidoping has no effective answer since testing is missing most of those athletes.
Because mathematics is one of my strengths, it was quite easy for me to accept the paper, with an abundancy of mathematical models and equations, in its entirety, including the discussion of all the limitations and potential sources of "distortion" (their word) -- not to mention the survey data results for the alternate SSC model estimating 21.4%, or 1 in 5.
Nothing in the paper says "more than 1 in 2 athletes" at the 2011 WCA. A lower "lower bound" figure could be less than 1 in 4 (see Appendix 1, Table 4, 50% fastest responses removed, 95% confidence, estimating 24.1%), as the researchers argued that these values are stable. But even that lower figure is not a strict lower bound, as it is still subject to all of the same limitations and potential sources of "distortion".
Note from the subject of this thread, it's about "Doping Prevalence in Track" and most notably the figure of "43.6%". All of your other points you want to argue against no one, existed before the survey, and is independent of the study, and this thread.
Any appearance of inferiority is ultimately based on suggestions, maybes, and biased plausibility.
Maybe I need to highlight this here more. It almost directly proves that the model used in the first SSC paper is nonsense:
We also present a hypothetical scenario of 1000 athletes with a 55% prevalence of doping, where athletes report 75% of the actual number of “true” statements. In this scenario, the P-model would yield only an 11.2% estimated prevalence of doping, rather than the actual number of 55%—again suggesting that this model may not be the best for analyzing SSC data.
In brief: the test case had 55% dopers, the SSC "analysis" (P-model) estimated just 11.2%!
That is just terrible. Terrible^3.
And it is not surprising either, considering that the same "analysis" estimated just 8.6% of supplement users at Daegu 2011, which is just ridiculously low if you ever talk to any elite runner. For example, people were quite surprised here when Houlihan claimed that she "only" used 4 - 5 different supplements. If the SSC "analysis" were correct, over 90% of the elite runners wouldn't use any supplements at all, and we'd be shocked to hear Houlihan taking several...
I would suggest to stop taking the SSC "analysis" seriously. But, you do you.
Nothing in the paper says "more than 1 in 2 athletes" at the 2011 WCA.
You are all over the place. On the same page here in the same thread, you just wrote:
the study itself presented no less than 39 potential estimates, in an attempt to eliminate potential distortions, ranging from 8.9% to 73.1%.
And on page 1 you wrote:
When you look at the extreme lowest to highest published values given for 95% confidence intervals, the values for doping estimates range from 9% to 47%.
So according to you, some published estimates in the study itself are well above 47% and even well above "1 in 2 athletes", all the way up to 73%, but according to you, 47% is the highest published value, and nothing in the paper says "more than 1 in 2 athletes".
Is "rekrunner" a group account for a bunch of WA employees? And you don't read each others posts?
Any appearance of inferiority is ultimately based on suggestions, maybes, and biased plausibility.
Maybe I need to highlight this here more. It almost directly proves that the model used in the first SSC paper is nonsense:
We also present a hypothetical scenario of 1000 athletes with a 55% prevalence of doping, where athletes report 75% of the actual number of “true” statements. In this scenario, the P-model would yield only an 11.2% estimated prevalence of doping, rather than the actual number of 55%—again suggesting that this model may not be the best for analyzing SSC data.
In brief: the test case had 55% dopers, the SSC "analysis" (P-model) estimated just 11.2%!
That is just terrible. Terrible^3.
And it is not surprising either, considering that the same "analysis" estimated just 8.6% of supplement users at Daegu 2011, which is just ridiculously low if you ever talk to any elite runner. For example, people were quite surprised here when Houlihan claimed that she "only" used 4 - 5 different supplements. If the SSC "analysis" were correct, over 90% of the elite runners wouldn't use any supplements at all, and we'd be shocked to hear Houlihan taking several...
I would suggest to stop taking the SSC "analysis" seriously. But, you do you.
The "suggestions", "may be", and "plausibility" came from your quotes.
The flaw in this critique can be traced to the false assumption introduced when creating the "A-model", that a new fudge-factor "p" is needed to arrive at true prevalence, when in fact it is only introduced to arrive at UQM estimates.
To summarize, the "P-model" (a subset of SSC), calls non-compliance "cheating". The "A-model" describes two categories of "cheating", one it calls "cheating" if it is deliberate, and one called "cognitive limitation" if it is accidental, and then assumes the P-model doesn't account for "cognitive limitation". This is a false conclusion, as the "P-model" accounts for all non-compliance, whether you call all of it "cheating" or want to split it up and name multiple sub-mechanisms of "cheating".
The problem with their hypothetical exercise of 1000 athletes with 55% prevalence, is that they ran the "P-model" on the "A-model" output, generated with the new fudge-factor "p". The result is also missing the percentage of non-compliance, i.e. "cheating". The fit is not good, because the distribution of the non-compliance differs between the "A-model" and the "P-model".
Regarding "supplements", no supplement data was collected at Daegu. That data comes from the Pan Arab Games -- a kind of Arabian Olympics composed of many sports besides running.
Of course SSC also has limitations. While the "P-model" implements just one non-compliance hypothesis, SSC actually evaluates many hypotheses in parallel, and chooses the best-fitting one. The set of hypotheses used in SSC for these events was limited, and could be augmented with more hypotheses. Our UQM researchers here picked a new non-compliance hypothesis which doesn't match any of them -- hence the bad fit of the P-model on A-model data. Both UQM and SSC results are in doubt when non-compliance is high.
Whether we take SSC seriously or not, this does nothing to confirm that UQM results are accurate, or superior to SSC.
Wasn't there some study that asked a hypothetical that if they could take some new drug that would guarantee a gold medal in 5 years but they'd be dead in 10 years .. would they take it? And like 80% said they would.
This study may have been done multiple times, but the example I’m aware of is quite old. My college track coach, who was an elite athlete in the 1970s, talked about having participated.
Nothing in the paper says "more than 1 in 2 athletes" at the 2011 WCA.
You are all over the place. On the same page here in the same thread, you just wrote:
the study itself presented no less than 39 potential estimates, in an attempt to eliminate potential distortions, ranging from 8.9% to 73.1%.
And on page 1 you wrote:
When you look at the extreme lowest to highest published values given for 95% confidence intervals, the values for doping estimates range from 9% to 47%.
So according to you, some published estimates in the study itself are well above 47% and even well above "1 in 2 athletes", all the way up to 73%, but according to you, 47% is the highest published value, and nothing in the paper says "more than 1 in 2 athletes".
Is "rekrunner" a group account for a bunch of WA employees? And you don't read each others posts?
"rekrunner" is just one guy.
Maybe I should qualify "within 95% confidence intervals". Note that none of the UQM confidence intervals considered the lack of confidence that would be introduced by the uncertainty of the many identified scenarios of survey non-compliance.
The researchers gave us two prevalence estimates, both of which are less than 1 in 2, and one is less than 1 in 3. In addition, they gave us many theoretical tables relative to the higher estimate. One can imagine publishing a revised set of tables relative to the lower estimate.
The Armstronglivs claim was "analysis of the survey figures suggest the extent of doping at championship level in athletic sports could be more than 1 in 2 athletes".
But that claim is wrong. The analysis of the survey figures did not suggest that. Any suggestion that it could be more than 1 in 2 comes from theoretical assumptions of sufficiently high non-compliance in selected non-compliance scenarios, rather than resulting from any "analysis of the survey figures".
On the contrary, what we did see from the sensitivity "analysis of the survey figures" is that the fastest responses contain a significant positive bias which is difficult to explain if the survey results are to be considered accurate -- greatly undermining any suggestion of "could be more than 1 in 2", and requiring even higher theoretical assumptions to counter-act the empirically observed positive bias, and then some.
All the figures from the survey are subject to interpretation - which is how you are trying to argue the lowest possible percentage will be doping. But there are credible interpretations that suggest doping could be even higher than 1 in 2 athletes. But whether it is less than that - say, 1 in 3 - the overall picture is the same, of a sport absolutely rife with doping and in which only a fraction of the dopers are caught. That is why doping continues as it does.
On the contrary, what we did see from the sensitivity "analysis of the survey figures" is that the fastest responses contain a significant positive bias which is difficult to explain if the survey results are to be considered accurate -- greatly undermining any suggestion of "could be more than 1 in 2", and requiring even higher theoretical assumptions to counter-act the empirically observed positive bias, and then some.
ChatGPT explains the "significant positive bias" - liars respond slower. Try removing the slowest responders, less honest responders, as a test - what do you get then?
ChatGPT: Do people respond slower or faster when contemplating to lie?
Great question — and it’s one that cognitive psychology and neuroscience have studied closely. Here's the core finding: 🧠 People generally respond slower when contemplating a lie. 🔹 Why? Lying usually involves higher cognitive load than telling the truth. To lie successfully, someone must: Inhibit the truthful response (which often comes automatically), Construct a plausible alternative, Monitor the lie’s consistency with known facts, Manage emotional stress or fear of being caught. This extra mental work takes time — often just fractions of a second, but measurable.
All the figures from the survey are subject to interpretation - which is how you are trying to argue the lowest possible percentage will be doping. But there are credible interpretations that suggest doping could be even higher than 1 in 2 athletes. But whether it is less than that - say, 1 in 3 - the overall picture is the same, of a sport absolutely rife with doping and in which only a fraction of the dopers are caught. That is why doping continues as it does.
They are also all subject to some combination of the at least the 8 sources of distortion from the identified survey non-compliance scenarios.
The UQM paper asks us to consider the nearly 1 in 2, and the less than 1 in 3. Outside UQM, we can also consider the biological estimate of 1 in 7, for blood doping. Or we can find back Howman's comment about cycling that doping is more than 1 in 10. This wide range of interpretations paint very different pictures about how rife doping is, and how large the fraction is that get's caught.
You keep falling back to the same keywords, like "credible" to speculatively suggest a possibility that could be. The UQM researchers liked to use the word "plausible", conflating plausibility with likelihood. These adjectives don't belong in the realm of science, but rather in the realm of mythology.
ChatGPT explains the "significant positive bias" - liars respond slower. Try removing the slowest responders, less honest responders, as a test - what do you get then?
ChatGPT: Do people respond slower or faster when contemplating to lie?
Great question — and it’s one that cognitive psychology and neuroscience have studied closely. Here's the core finding: 🧠 People generally respond slower when contemplating a lie. 🔹 Why? Lying usually involves higher cognitive load than telling the truth. To lie successfully, someone must: Inhibit the truthful response (which often comes automatically), Construct a plausible alternative, Monitor the lie’s consistency with known facts, Manage emotional stress or fear of being caught. This extra mental work takes time — often just fractions of a second, but measurable.
The UQM researchers justified excluding the fastest responses as hasty, lacking any contemplation. They weren't lies so much as an unwillingness to be bothered to read and understand the strategy of anonymity and the protection it affords, and comply with the given survey instructions.
But I agree with you -- that doesn't mean the slowest 70% are honest responses. A significant but stable positive bias from "automatic" responding, can still be present throughout the remaining 70%.