Power Data Estimates for the climbing stages

[acoggan]

Check coggan's post history.
Said something that even if a rider did that he would not believe it was doping.

No, I did not. I said that it wouldn't be proof of doping.

 

[Le breton]

No, I did not. I said that it wouldn't be proof of doping.

10 kW: Waouh. His mechanical efficiency better be close to 100% or else he will burst in flame pretty quickly.

Do you have a fire extinguisher in your lab if such an alien ever walks in?

Alien Lim had lots of water to prevent Fluid Landis from catching fire.

 

[Bumeington]

I can't see the link, but I'm glad to see at least one person is asking the right question.

All of this "wind speed", "power calculation" stuff is complete nonsense. Unless there's a rider in that front group with known weight and a well-calibrated power meter (two very big "ifs"), determining w/kg is going to be fraught with enough error that it's completely useless to make any determinations regarding doping. This is particularly true on a climb like Ventoux that has high, variable winds and is exposed. Plus, strategy and race situation always play a part.

The real question is "how much more power did Froome have to put out compared to the other guys". If Ferrari's calculations are correct, then the answer is "a lot". Enough of a difference to be plenty suspicious, particularly given Froome's pedigree. >10% over the next non-Sky rider isn't exactly "marginal", it's the advantage you'd get with a very well-orchestrated doping program.

I don't understand why Ferrari does this square speeds thing, he does it for the climbs and the tt's as well.

According to my interpretation of basic physics, I'd expect power proportional to speed^(~2.85) at typical tt speeds with a light wind and power/weight proportional to speed^(~1.25) on a typical alpine climb.

Both these numbers 1.25 and 2.85 are a balance between air resistance which is Power~speed^3 and rolling resistance power ~ speed and gravity where power/weight ~ speed.

Using these numbers I get Froome as 16% more power (well actually power/cda) than valverde in the TT, 6% more power/weight than valverde in ax-3-domines, which would fit almost perfectly with Froome as 67kg and Valverde as 61kg.

 

[Almeisan]

No, I did not. I said that it wouldn't be proof of doping.

So it's proof of what? Proof of not even being biological? So we way passed the 'proof of doping' barrier. So where is it?

You deny there is a barrier to human performance or what. Or you just refuse to consider the concept.

Yeah, this isn't math so 'proof' means something different. He was probably equivocating the meaning of 'proof' to muddy the discussion and refuse to give a straight answer.
No amount of positive doping tests will be 'proof' of doping. It will all be evidence.


There have been athletes who we know for a fact know doped and WADA 'proved' that they did not dope.
There have been athletes who we know for a fact did not dope and WADA 'proved' they did dope.
Sure WADA are doing a decent job with the resources they have. But some people like to know the facts. And we know that in most cases dopers get away with their fraud.

So people try to do some wattage estimations, for fun and because of doping suspicious and to confirm them or alleviate them. This Coggan muddying up and derailing isn't constructive to this.

 

[131313]

So lets say that Froome produced 10% more power (expressed in w/kg) than Contador. So f@#king what? What does that prove? We already know he finished 2min ahead or whatever.

1) why are you always so angry, and 2) who said it "proved" anything? Why don't you just read the rest of my post instead of pounding your fist on the desk like a petulant child?

 

[131313]

Allen Lim, sportscientists.com, Vayer, Lemond...the list is quite long, actually.

I think you're mis-characterizing their positions a bit, which the exception of Vayer, who's an idiot who can't do basic math and has some clear, strong biases. "suggestive of doping" doesn't equal "proof".

 

[acoggan]

I think you're mis-characterizing their positions a bit, which the exception of Vayer, who's an idiot who can't do basic math and has some clear, strong biases. "suggestive of doping" doesn't equal "proof".

I don't think so. For example, Lim came out quite clearly on precisely where he thought the "doping line" should be drawn (even though, ironically, estimates of his employer's, i.e., Armstrong's, power would put him above that line). The same is true for Tucker and Dugas, at least a couple of years ago (they have backed off significantly on their stance more recently).

As for Lemond, I don't recall him ever stating a specific value, but he didn't believe that Contador's performance on Verbier (sp?) was believable, and has long argued that power data can and should be used to detect doping.

 

[acoggan]

So it's proof of what? Proof of not even being biological? So we way passed the 'proof of doping' barrier. So where is it?

The 'doping line' in terms of power output? I don't really know - that's my point.

 

[dansmith]

apparently, this has been a sticking point with others too...getting at the real data...

"LeMond calls for Froome, others to release power data"

http://velonews.competitor.com/2013...or-froome-others-to-release-power-data_295268

"Q&A: Brailsford on why Froome won’t release power data"

http://velonews.competitor.com/2013...-on-why-froome-wont-release-power-data_293771

"Transparency & Grand Tour Contenders"

http://bikepure.org/2013/06/transparency-grand-tour-contenders/

"Watts the Story to Tour de France Glory?"

http://bikepure.org/2013/06/watts-t...ory-report-into-doping-in-the-tour-de-france/

 

[131313]

I don't think so. For example, Lim came out quite clearly on precisely where he thought the "doping line" should be drawn (even though, ironically, estimates of his employer's, i.e., Armstrong's, power would put him above that line). The same is true for Tucker and Dugas, at least a couple of years ago (they have backed off significantly on their stance more recently).

As for Lemond, I don't recall him ever stating a specific value, but he didn't believe that Contador's performance on Verbier (sp?) was believable, and has long argued that power data can and should be used to detect doping.

OK, with Lim I'll admit when he speaks I don't listen. He's always been a fraud and huckster in my book. The SS guys most recent comments are what I'm going on, and they echo what LeMond has said, or at least how I'm interpreting what he's said. I think "using power data to detect doping" is exactly what he's suggested, and I think it's a reasonable idea. Not to convict or suspend someone, but to put in action some specific targeted testing. If someone shows a 10% increase in power from one year to the next, I think they're deserving of extra scrutiny. That's a far cry from sanctioning someone because they ride over a certain w/kg, and as far as I can tell they aren't calling for that. If they are I'll be right next to you in disagreement.

Of course, this only works if those in charge of testing are trying to actually catch the riders who are doping, and right now that seems to be an open question.

 

[Alex Simmons/RST]

apparently, this has been a sticking point with others too...getting at the real data...

"LeMond calls for Froome, others to release power data"

http://velonews.competitor.com/2013...or-froome-others-to-release-power-data_295268

"Q&A: Brailsford on why Froome won’t release power data"

http://velonews.competitor.com/2013...-on-why-froome-wont-release-power-data_293771

"Transparency & Grand Tour Contenders"

http://bikepure.org/2013/06/transparency-grand-tour-contenders/

"Watts the Story to Tour de France Glory?"

http://bikepure.org/2013/06/watts-t...ory-report-into-doping-in-the-tour-de-france/

OK, let's for a minute assume that tomorrow the riders publish their power data.

Then what?

Here are my thoughts.

 

[Merckx index]

OK, let's for a minute assume that tomorrow the riders publish their power data.

Then what?

Here are my thoughts.

Just where is the doping power plausibility line? Can we really assign such a line? Is 6.2W/kg for an hour proof? 6.3? 6.4? 6.41?
A: In reality we simply can't put a clean line in the sand. The line for each rider may be different, and the line may vary depending on context. How long was the effort? When did it occur? What were the environmental conditions? How steep was the climb? Was it solo or with others? Was it a consistent effort or variable? Who responds better to doping?

I think you’re exaggerating the problems. There is a line beyond which we can say with very high certainty that doping must have occurred. There are ways of controlling for the effect of time on power, definitely, there are well known if not exact relationships between the two. The effects of the other factors can also be estimated. Moreover, as I have suggested repeatedly before, a climb could be put in a GT, year after year, that was specifically designed to minimize confounding factors, such as environmental conditions, changes in gradient, etc., and if it were a TT, it would also avoid problems of drafting, tactics, etc. Hell, if cycling were serious, they could have a “stage” in which riders just pedaled a machine, with the identical artificial “gradient” or work load for each rider. That would eliminate all variables. Sure, it would be a funny kind of stage, not spectator friendly, and purists would complain no end that it wasn’t real bike racing. Well, tough ****, if one is really serious about getting at the power riders are putting out, it could be done and would be done. And in such a way that riders couldn’t fake their output without seriously damaging their chances of winning the GT.

Will it prove riders aren't doping?
Of course not. Since it assumes there is an arbitrary upper power limit for doping to be confirmed, it does nothing to pick up any doping by riders who are below whatever that arbitrary limit is. No green jersey contender for instance is going to out ride the GC contenders on major cols. Hence such data only serves to tell us what we already know, i.e. a handful of riders finished ahead of their competition on the mountain top finishes.

By that logic, a positive test won’t prove riders are doping, either, since a test also frequently assumes an arbitrary limit for confirmation. Most doping tests are based on probabilities, which is also what setting an arbitrary power number is all about. E.g., much of the controversy over the Landis case revolved around whether the probability of his isotope ratios was low enough to conclude doping. You can have legitimate debate about what the criterion should be, but you can’t have debate over whether there is in a fact a criterion.

The fact that many riders might be doping and not meet the criterion because they aren’t climbers is not relevant. No one is saying that power data should be used to sanction someone (at least I’m not). We’re just saying it would give us a better idea. So the fact that there might be a lot of false negatives doesn’t matter. The question is whether we can identify some real positives. It’s well known that there is a tradeoff between false positives and false negatives, so if you’re willing to allow a lot of the latter, the positives you do identify can be at any arbitrary level of significance. This is another way of stating that there has to be a line above which we can indeed say with any degree of confidence we want that a rider must be doped.

Can power data be manipulated?
A: Yes, of course it can. Accidentally, inadvertently or deliberately. So then we'll have those on the conspiracy trail of a new doping avoidance technique of "data doping". Since we already know the amount of slop in power estimates from other methods, then fiddling with the numbers means no-one can really know if numbers are fiddled or not. There are of course forensic data analysis techniques that can identify some examples of that, but only if crude data manipulation methods are used. If riders and their support people are clever enough to manipulate blood to avoid detection, I'm pretty sure they'll be able to work out how to manipulate data to avoid detection.

What would it cost to run such data collection in an independent manner, and free from possible manipulation?
A: Millions of dollars. Think about the number of bikes in the ProTour, the need to carefully calibrate say 1,000 SRMs, to have non-tamperable data loggers, to ensure all riders correctly perform zero-offset checks before and during races. The data collection process. Staff to manage this. Millions of dollars that perhaps would be better directed at improving doping control processes, technology, reducing testing costs, and simply performing more tests and more frequently testing in and out of competition.

Again, I think you’re exaggerating the problem. We don’t need one thousand SRMs, a few dozen would cover the riders and stages of most interest. How many riders in a GT are genuine contenders for a jersey, particularly the GC? How many stages are critical? That would be the main focus, at least in the begining. Moreover, much of the data could be checked by comparing outputs from riders who finished together or close together on specific climbs, for example.

What is there to lose? Drug testing is next to useless. It probably catches less than 10% of those doping at any particular time. I think this should be given a try. I’m not saying the power data have to be publicly released, they could be restricted to an independent panel, but why not give this a try?

 

[131313]

OK, let's for a minute assume that tomorrow the riders publish their power data.

Then what?

Here are my thoughts.

Alex, with respect, I think your thoughts are pretty misguided. It seems to be an overreaction to some of the nutball conspiracies you read on here, both in the content of what you're saying and how you say it.

First off you're right, if riders release power data there will be questions about its integrity. That's not "conspiracy theory", or automatically indicative of "confirmation bias", that's healthy skepticism. If a rider climbs Ventoux at record time @ 5 w/kg, there are going to be questions, as there should. And it won't surprise me if there are some wonky files, intentional or not. But guess what? If ALL of the riders release their power data, it will provide enough information that it will be easier to determine the actual power riders are putting out, unless of course they're all conspiring together to doctor their power files in the same manner. It will simply add additional data points to the estimations.

Secondly, I think you're off base saying that it holds no usefulness. It could be a piece of the puzzle, just like the biopassport data. Of course, as I mentioned before this is only useful data if those evaluating it are interested in catching dopers. If a rider's threshold power increases 10-15% after several years of being within a 2-3% range, don't you think that rider should be subject to additional out-of-competition controls with more complete panels done? I think so.

The alternative of course, and the one you seem to be suggesting, is that we simply look at nothing besides drug testing and say "mission accomplished". That seems a bit too "head-in-the-sand" for my tastes, and it sings a familiar refrain. Bottom line, I see some usefulness in evaluating performance and comparing it to past performances.

 

[lean mean &green]

good posts MI and 131313.

we also don't need a hundred million bazillion $$$ (simmons estimate ) right away to start the initiative. the use of SRM data could start small. initially, maybe the best placed rider on GC for each team must produce data for the day. gradually most or all riders in a grand tour, eventually all protour athletes, etc.

it's simply another data point added to the biopassport. not sure what all the fuss is about.

there are a couple of "experts" in this thread/forum who spend a lot of time trying to look superior to "laypersons" and very little time looking for real solutions.

 

[Alex Simmons/RST]

I think you’re exaggerating the problems. There is a line beyond which we can say with very high certainty that doping must have occurred. There are ways of controlling for the effect of time on power, definitely, there are well known if not exact relationships between the two.

Thanks for reading/responding. I figured it'd provoke a bit of a discussion.

Not sure I'm exaggerating problems, more that I just don't see this being a solution to combat doping that many out there would seem to think.

As for standard stages/tests, well I just can't see such events incorporating such tests. Only place to control for the variables is a lab. That's just not going to happen. This is a race.

By that logic, a positive test won’t prove riders are doping, either, since a test also frequently assumes an arbitrary limit for confirmation. Most doping tests are based on probabilities,

There is a big difference between a test that is used to declare whether a substance actually is detected in blood/urine, and a power meter file which simple records what the rider did performance wise. The power meter file does not tell us whether a rider used a prohibited substance or method, just as climbing ADH in 40-minutes doesn't either.

The fact that many riders might be doping and not meet the criterion because they aren’t climbers is not relevant. No one is saying that power data should be used to sanction someone (at least I’m not). We’re just saying it would give us a better idea. So the fact that there might be a lot of false negatives doesn’t matter. The question is whether we can identify some real positives. It’s well known that there is a tradeoff between false positives and false negatives, so if you’re willing to allow a lot of the latter, the positives you do identify can be at any arbitrary level of significance. This is another way of stating that there has to be a line above which we can indeed say with any degree of confidence we want that a rider must be doped.

I'm confused with your terminology. You are saying we can't sanction based on power data but it can be sufficiently indicative of being positive for doping. If it's sufficiently indicative, then surely sanction must follow? Else it's pointless.

In what way does the data tell us who should be targeted that we don't already know by the fact that they are winning/elite/professional bike riders?

This is another way of stating that there has to be a line above which we can indeed say with any degree of confidence we want that a rider must be doped.

OK, so what's the line? And why?
Is the line the same for say Mark Cavendish as it is for Alberto Contador or <insert any rider name here>?
If not, why not? How would you determine this?

Again, I think you’re exaggerating the problem. We don’t need one thousand SRMs, a few dozen would cover the riders and stages of most interest. How many riders in a GT are genuine contenders for a jersey, particularly the GC? How many stages are critical? That would be the main focus, at least in the begining. Moreover, much of the data could be checked by comparing outputs from riders who finished together or close together on specific climbs, for example.

Perhaps, but if we are going to implement a doping control regime, should it not apply equally to all?

Is it OK to allow the helpers to get away with it, after towing their leader all day so they can deal with the finale?

And what if a rider has a "mechanical" and changes bikes? Are we going to stop the race so an independent referee can validate the new power meter's zero offset and slope setting are correct (assuming you have a power meter on the spare)? Do we ban using a team mate's bike if that became necessary?

Practical arguments exist for good reason. If you are going to implement something, it needs to be workable.

Then there is the context argument that you allude to earlier (i.e. being able to account for things like length of stage etc when deciding on what the line in the sand is). Well if you don't use the data to account for the rider's workload through the whole race from Day 1, indeed for the months leading up to the race, then doesn't that sort of negate the ability to account for such things?

Some voices on this point have all said that they can believe e.g. 6W/kg in some situations but not in others. So how is that going to be sensibly accounted for if you are not collecting the actual data to place that 6W/kg effort in its proper context?

What is there to lose? Drug testing is next to useless. It probably catches less than 10% of those doping at any particular time. I think this should be given a try. I’m not saying the power data have to be publicly released, they could be restricted to an independent panel, but why not give this a try?

Hey, I'm all for release of data.

My main point is I think people are expecting too much from such data and it's not really going to add much to the task of finding out who is doping.

What have we to lose? Well for the data that already exists, nothing, but I don't think we really particularly gain anything either. It will simply add to the pre-existing confirmation biases. e.g. if you are pre-conceived to the notion that a certain rider is doping but all we see is 5.9W/kg from them, then all such people will do is reset what's "plausible".

If however it became a formal requirement to have power data that was managed independently, then I do think we do lose, as the resources to do that have to come from somewhere, and that would likely be at the expense of improved actual doping control. Would the resources not be better directed at improved detection methods, increased OOC testing etc?

 

[Alex Simmons/RST]

Alex, with respect, I think your thoughts are pretty misguided. It seems to be an overreaction to some of the nutball conspiracies you read on here, both in the content of what you're saying and how you say it.

First off you're right, if riders release power data there will be questions about its integrity. That's not "conspiracy theory", or automatically indicative of "confirmation bias", that's healthy skepticism. If a rider climbs Ventoux at record time @ 5 w/kg, there are going to be questions, as there should. And it won't surprise me if there are some wonky files, intentional or not. But guess what? If ALL of the riders release their power data, it will provide enough information that it will be easier to determine the actual power riders are putting out, unless of course they're all conspiring together to doctor their power files in the same manner. It will simply add additional data points to the estimations.

Secondly, I think you're off base saying that it holds no usefulness. It could be a piece of the puzzle, just like the biopassport data. Of course, as I mentioned before this is only useful data if those evaluating it are interested in catching dopers. If a rider's threshold power increases 10-15% after several years of being within a 2-3% range, don't you think that rider should be subject to additional out-of-competition controls with more complete panels done? I think so.

The alternative of course, and the one you seem to be suggesting, is that we simply look at nothing besides drug testing and say "mission accomplished". That seems a bit too "head-in-the-sand" for my tastes, and it sings a familiar refrain. Bottom line, I see some usefulness in evaluating performance and comparing it to past performances.

Well someone has to respond to the nonsense out there.
Anti-vaxxers still get air time and they need to be put in their place, else chaos will reign

But the crux of your suggestion is that we need data from the majority, and we need it for a majority of their riding in order to make sense of it. So I suppose that then brings us back to the "bazillion $$$" problem

As for how to determine if a rider should be subjected to increased OOC controls, sure perhaps such a jump in power might be a trigger, but then it will show up in race performances anyway, and since they are at the elite level they should be subject to such testing anyway.

Or don't we think a panel could work that out from simply monitoring actual race performance?

 

[armchairclimber]

It's really good to see intelligent posts in here.

 

[Le breton]

It's really good to see intelligent posts in here.

Yes, I agree. Hope I have time to come back to it later. Thanks Alex, 131313, MI.

 

[Alex Simmons/RST]

Speaking of intelligent posts, have a read of this item which puts the pVAM model under some sound physiological and physics scrutiny:

http://djconnel.blogspot.com/2013/07/pvam-and-critical-power-model.html

 

[Ferminal]

Look at any set of data, it's pretty impossible to see a relationship between time and power unless you're comparing Mur de Huys to Stelvios.

 

[Fearless Greg Lemond]

Speaking of intelligent posts, have a read of this item which puts the pVAM model under some sound physiological and physics scrutiny:

http://djconnel.blogspot.com/2013/07/pvam-and-critical-power-model.html

It is funny how you, Coggins, Krebs are bending in all sorts of ways to 'debunk' the w/k calculations. Do you guys have a vested interest or something?

The calculations are not 100%, the error margin on SRM's are documented at most 2% for a number of riders. Still within the 2.5% standard deviation I would say. But, they are not a gospel. In the Vayer magazine 21 Counts the authors clearly state for example the Ventoux stage measurement after Chalet Reynard has a 5% margin of error due to wind conditions, that is too much. The part between Esteve and Reynard however has a 2% margin.

Or are you guys now saying SRM is unreliable?

 

[meat puppet]

It is funny how you, Coggins, Krebs are bending in all sorts of ways to 'debunk' the w/k calculations. Do you guys have a vested interest or something?

wouldn't go that far, TBH. the w/kg measure will probably be popularised anyhow, even among casual fans. it is a ballpark figure and as such quite sound. i take it the people plotting the statistics know they are on the mark in what they are doing. ie. not shooting down individual dopers, but establishing baselines.

also, it doesn't hurt to look at the models critically, even if the critical input isn't necessarily constructive. quite the opposite, I dont think any of the "pseudo scientists" (brailsford 2013) claim that their current models are the be all end all in this but rather welcome criticism and input to improve them. which is exactly what intellectually honest persons would do at this point, no?

anyhow, doc plotted some power curves from key riders just recently. dont look exactly like curves, tho. https://twitter.com/veloclinic

 

[Alex Simmons/RST]

dont look exactly like curves, tho.

Well if they are not all mean maximal power numbers then of course not.

 

[Fearless Greg Lemond]

also, it doesn't hurt to look at the models critically, even if the critical input isn't necessarily constructive. quite the opposite, I dont think any of the "pseudo scientists" (brailsford 2013) claim that their current models are the be all end all in this but rather welcome criticism and input to improve them. which is exactly what intellectually honest persons would do at this point, no?

That I agree with, of course.

 

[Alex Simmons/RST]

It is funny how you, Coggins, Krebs are bending in all sorts of ways to 'debunk' the w/k calculations. Do you guys have a vested interest or something?

The calculations are not 100%, the error margin on SRM's are documented at most 2% for a number of riders. Still within the 2.5% standard deviation I would say. But, they are not a gospel. In the Vayer magazine 21 Counts the authors clearly state for example the Ventoux stage measurement after Chalet Reynard has a 5% margin of error due to wind conditions, that is too much. The part between Esteve and Reynard however has a 2% margin.

Or are you guys now saying SRM is unreliable?

I can't speak for the others.

I don't think I'm debunking W/kg calculations, just encouraging people to think critically about how such information is arrived at and also what (IMO) we can and can't interpret from the information.