Welcome to the Science of Sport, where we bring you the second, third, and fourth level of analysis you will not find anywhere else.

Be it doping in sport, hot topics like Caster Semenya or Oscar Pistorius, or the dehydration myth, we try to translate the science behind sports and sports performance.

Consider a donation if you like what you see here!

Did you know?
We published The Runner's Body in May 2009. With an average 4.4/5 stars on Amazon.com, it has been receiving positive reviews from runners and non-runners alike.

Available for the Kindle and also in the traditional paper back. It will make a great gift for the runners you know, and helps support our work here on The Science of Sport.

Monday, July 26, 2010

Tour power output reflections

Looking back on leTour - the post collection

Thank you for visiting The Science of Sport. Over the past few weeks, we've followed and attempted to analyse the performances of the very best cyclists in the world, and at worst, it's created some great discussion and back-and-forth.  At best, it's shown that cycling may just be heading in the right direction in its fight against doping.

Earlier today, Greg Lemond mentioned our analysis in his blog at Cycling News under the title "data of optimism?" and I certainly share that sentiment.  So for those arriving "late", below are the links to the three analysis we've done on the power outputs, courtesy data provided by SRM and Training Peaks.

Post 1: Power outputs from the Alps and Pyrenees
Post 2: The Col du Tourmalet - the showdown at 6W/kg
Post 3: Resolving discrepancies in the Tourmalet numbers

I'd encourage you to also read the comments, where you have really improved the overall quality of the debate with your own calculations and questions.

One of the big talking points in all these analyses is the issue of whether a performance is proof of doping.  Of course, the answer is no.  There are too many assumptions in the calculation of physiological implications of a given performance for it to be "proof".  Also, things like tactics and weather and preceding stages affect a rider's ability to produce a given power output.  However, when looked at in context and when those assumptions are "controlled" in order to create a 'best-case scenario', the picture is still, I believe, telling, and that is what the above posts are about.  There comes a point at which the principle adds value.

Of particular interest given the debate before the Tour, is that not a single longer climb hit the power outputs that we've become accustomed to seeing in 90s and 2000s.  Nor have they hit what we debated pre-Tour as the "suspect" power values of greater than 6.2, 6.3, 6.4 W/kg.

And while the 6.2 W/kg number got a lot of people riled, I really think it's telling that the very best climbers, with the highest level of motivation (on the Tourmalet) failed to hit those power outputs.  Re that number - in a debate about "unrealistic performances", you have to commit to a value, even if only to illustrate a point.  It does not mean this number separates the world into light and dark.

Even Contador and Schleck on the Tourmalet, in what was an absolute 'limit' performance, just touched 6W/kg as an average, and appear to have dropped right down towards the end of the climb (see post 3 above).  To me, this largely validates the physiological principle that says that for every performance, there is a physiological 'cost' and at some point, the 'cost' becomes an indication of doping.  In the words of Lemond, the performance becomes "believable".

There is no dividing line in the sand, no specific point at which you can say "got you".  A rider at 6 W/kg may be doping, and one at 6.2W/kg (depending on the situation) may not, but there is a theory underpinning it and the change in this year's Tour is a positive sign, leading to the hypothesis made in those posts and by Lemond.

It's been a super Tour, with great individual performances on stages, and the confirmation of a rivalry between Contador and Schleck that will hopefully put cycling in the news for the right reasons.  And hopefully, it's also produced a step in the right direction for the sport.  Bring on 2011, hopefully a mountain time-trial, and another super-tight race!


Donate and support The Science of Sport

Thank you for taking time to read and follow our coverage of the Tour de France here at The Science of Sport. We hope you've enjoyed the insights we've tried to provide! Your time and energies are greatly appreciated!

We run this site as a "labour of love" and will continue to do so, but of course, any support is greatly appreciated. So, if you would like to donate and support our continued efforts to bring you the insights and analysis, please consider donating to our site. You can do so in one of two ways:

  1. If you are reading this on our site, then simply scroll up to the top of the page, where you will find the DONATE button at the top right of the page.
  2. If you are reading this in our email newsletter, click here to be taken to our site, where you will see the DONATE button at the top right of our homepage.
Thank you once again, and keep on reading!


CowPunk said...

The problem is you assume a rider is riding is hard as they can at all times which isn't true.

So actually you're trying to make a point that cannot be made unless you are carrying out an experiment in controlled conditions with riders going all out the entire time in addition to a control group. In fact even if you looked a the TT results, you couldn't be certain all the riders were going all out unless you paid them to do so.

What you really have is just a bunch of useless numbers that fools like LemonD pick up on for getting PR and ruining other peoples careers and reputations in the process.

Ross Tucker and Jonathan Dugas said...

To Cowpunk:

You raise a very valid point in an extreme way. Of course, they're not always riding maximally. We've acknowledged that dozens of times in the last few weeks - this is why I really implored people to read the posts and the comments, and perhaps you did.

But this assumption does not mean that they are useless numbers. The reality is that the numbers now are lower than the numbers of the 90s and 2000s. They are also lower than the numbers that sound physiological principles would hypothesize to be in a gray area. So the comparison, the longitudinal evolution of performances is telling. Unless of course you're assuming that nobody in the 90s and 2000s held back? That in that era, everyone was going as hard as they could?

Of course not, and so as the body of data is built up, the ability of this kind of observational study to create value and meaning improves.

Note that i'm agreeing with you, but you're making the point with some hostility (presumably against Lemond) which is unwarranted - we've explained this error many times. And no, you're wrong - the only science is not the controlled condition one. Your view of the field is far too narrow.


Gene said...

Ross, I think your academic queasiness - a truly infectious disease - is showing. As I understand you main point, I wouldn't agree with CowPunk at all. You (and SRM) are comparing averages between riders covering the same or similar types of terrain. With sufficient data points, and I don't think many are needed in this type of case, the difference in average pace between one time period and another provide evidence to support the point about doping. At least about as good evidence as one is going to get without direct knowledge or possibly controlled lab studies (which suffer from not being real world).

Ross Tucker and Jonathan Dugas said...

hi Gene

Thank you for the comment, and even more for the donation (I am assuming).

You are quite right - perhaps I agreed to easily. Certainly, CowPunk is correct that this limitation exists.

I certainly don't agree that it's a bunch of useless numbers - I would spend time looking at it then! I think CowPunk has an issue with Lemond perhaps.

But yes, given sufficient numbers, the data become meaningful. And they confirm the pre-Tour hypothesis, which I also find important. And of course, there were stages, like the Tourmalet, where the effort was maximal, for everyone. So I'm confident in the data, and thanks for pointing that out!


Anonymous said...

Hi people , I don't want to annoy you , but I went back to stage 15 on Port de Bales . I checked the altitude where was Schleck after he started again after putting his chain on the bicycle with Google Map/Street View and also IGN map as suggested by Frederic Portoleau. He was at 1600 m above sea level. Then I looked at the movie on French TV ( sport.francetv.fr ) because I don't have the record. It seems to me that he started again at 16.43.40 and he arrived at the top at 16.48.44 . If you have the record you can measure your own time. So he employ 5 minutes and 4 seconds to climb 155 m at 1836 meter/hour. In the next days I will try to check the final part of the Tourmalet to see if there is some decrease from Schleck /Contador as you are supposing , but it will be hard because there is a lot of fog.


Jonas said...

I just wanted to thank you guys for running this great site. I just recently discovered it, but I have eagerly read the posts during this year's TdF. As an MD doing a bit of (clinical) research myself I really appreciate your approach: Evidence-based, transparent, and reproducible, and the interpretations of the data are insightful and clearly articulated. I also find the comments and replies very instructive.

Just to follow up on Pierre's post about that stage where Schleck's chain popped. Before the Tourmalet stage, I looked at the incident again at YouTube and it actually gave me some belief (in my case hope, as I "supported" Schleck) that he might actually be able to crack Contador on the Tourmalet.

The reason is this: From my unofficial count it seemed like Schleck actually gained about 20 seconds on Contador and Sanchez to the top after he got back on his bike following his chain problem. From the YouTube clip it seems that it was about 35 seconds after Contador passed him until Schleck was back on the bike, and at the summit he was only trailing Contador by roughly 15 seconds. In other words Schleck gained 20 seconds on Contador et al. Obviously, Schleck was probably super-motivated and angry at that point, and giving it everything he got, but it still seems like a lot given that Contador seemed to go full speed up the mountain as well.


Ron Wolf said...

A few reflections of my own.

First, a meta-reflection - I have enjoyed (in a sports science junkie kind of way) going along for the ride as you have developed your 6.2w/Kg limit hypothesis. At first I was quite skeptical, now I'm quite intrigued. Its satisfying to see how the various sources point to about the same conclusion. Also, quite interesting how you have used the give and take of blog discussion to hone this. Kudos!

Next, a latitudinal reflection - how does this 6.2w/Kg look in other sports? I think you discussed this at one point, but don't find it now. For instance what would the power output need to be to run a 2 hour marathon? Or to match the world record half-marathon? A world class 5k? And how about "whole body" sports such as x-country skiing where, because more muscle groups are used, O2 uptake is higher (at least that's what I recall). Would it be reasonable to expect 6.2w/Kg to be exceeded in these sports?

Finally, a future-reflection - what will it take for humans to exceed this seeming limit? I mean ethically, without doping at least. Is it only a matter of time before someone shows up who can out do this? Are new training methods on the horizon that will allow gifted athletes to exceed 6.2w/Kg? Or do you think this is simply, The Limit of Human Performance? Prior to genetically engineered humans that is....

CowPunk said...


Sorry to come across the wrong way. But Lemond is using your article as a point of reference to validate his accusations against other riders, potentially damaging their careers.


Frans Rutten said...

Jonas said

IMO there's no reason to read too much into Schleck's performance, after he resumed his race. He lost about 26 tot 27s with his mishap, but physiologically also had a 26s to 27s "rest". When, in full throttle, he was forced to stop, he had only a few seconds left to sustain this pace.

Alberto Contador was clear behind, when it happened, but decided then in a split second to make use of the confusion, and to ride on, but didn't do that with full speed all the way. In fact, he even waited for others.

Andy Schleck was faster on that last segment, no doubt, but I see no reason to state, that this segment proved, that Andy Schleck was the better climber in this Tour.

Jonas said...

Frans Rutten said

"Andy Schleck was faster on that last segment, no doubt, but I see no reason to state, that this segment proved, that Andy Schleck was the better climber in this Tour."

Yeah, I totally agree with that. There's no indication that Schleck was the better climber than Contador - case in point being the climb up Tourmalet. Contador had a couple of days where he looked vulnerable (like that stage in the Alps where Schleck got 10 secs on him and stage 15 where the chain popped), but then so did Schleck. If nothing else, I'd say that's another indication they were clean, not displaying the invincibility like Indurain and Armstrong did in the 90s and 00s.

Ross Tucker and Jonathan Dugas said...

Hi all

So many comments, I'll do my best!

To Pierre:

You're not annoying us, don't worry. Interesting re Schleck. I'm not all that surprised - those 5 minutes were perhaps among the hardest efforts of the whole Tour, given the situation at the time, so on that adrenaline, the need to close a gap, that was all out. What is more telling is the greater context of the whole climb, and as I recall, this was not all that quick - it took Horner 49:30, giving him a VAM of 1348 and power output of 5.2W/kg average (incidentally, the SRM gives 5.4W/kg - see post 1, so the SRM is a little higher than the VAM method, suggesting a headwind on the climb).

So for Contador, you're looking at a time of 46:00, a VAM of 1450m/h and a power output of 5.6W/kg. I think on this day, VAM is underestimating, so I suspect maybe 5.6W/kg to 5.8W/kg - that's not all that high, and so to ride the last 5 min at 1850m/h, not too bad.

To Jonas and Frans:

Agreed with both. This is, I guess, one of the great questions of the Tour - what if? I agree with you both that Schleck and COntador were basically inseparable on those long climbs. And I do think Contador did sit up a little on the Port be Bales.

And I agree with you Jonas on the 'vulnerability'. Alone, this is obviously very subjective, and maybe we're reading too much into it, but it is telling, not only of individuals, but of teams.

More to come...

Ross Tucker and Jonathan Dugas said...


To Ron:

Thank you, much appreciated. I must confess, I've also been intrigued to see how the 6.2W/kg issue evolved! When I did that first post, I had no idea it would prove so "divisive" (literally and figuratively), and with the benefit of hindsight, I'd have written that post a little differently, to be less specific about a number, more about a range.

But anyway, it has been interesting to track and the SRM data are really very interesting in this regard. It's a good thing that we've had Horner to act as a barometer for the front of the race, because he's conceding between 90sec and 3 minutes on longer climbs, and that's close enough that I think we can, with some certainty, dismiss the estimations re distances and altitude that sometimes affect VAM-type calculations.

Not that VAM has been useless - my calculations, and those by Thomas, Pierre and yourself, have actually been mighty close to the SRM values (within 0.2W/kg for Horner and SOrensen), that I'm pretty confident in saying that the top guys, based on VAM, were between 5.9W/kg and 6 W/kg.

Anyway, to answer your question - this is something I've started to look into, because it is the logical extension of this series. The series actually comes out of similar approaches to running - as you probably know, athletic performance has been modelled statiscially many times, to try to predict where world records will cease to be broken.

My thinking, physiologically, is that the reason this will happen is because at some point, the physiological requirement to run distance X in time Y is just not seen in humans. So logically, I take the same attitude to cycling.

Having said this, i don't know of a power output equivalent for running and cycling. That Joyner paper that Andrew Coggan mentioned over on CyclingNews looked at the metabolic/energetic demands, and it gave a figure of about 1:58 for the marathon. I guess one could attempt the same for shorter distances, but this gets very complex. Also, in running, the elasticity of muscle-tendon and the eccentric phase of the cycle makes it difficult because not all the work is done actively - some is 'passive' elastic recoil, and this is one reason why the east Africans have been surmised to have an advantage. Never been shown, but it's speculated that their tendons return more energy.

As I say, I'm going to look into it and see if power output can be estimated for running, to see if there is an equivalent to the now 'notorious' 6.2W/kg!

I will say that the principle will apply - at some speed, the energy cost (both in terms of oxygen required and energy burned) exceeds what is physiologically possible.

Doping would improve this and that's another series of posts altogether. Not sure if you ever saw this post though: http://www.sportsscientists.com/2007/11/effect-of-epo-on-performance-who.html

It gives an idea of how doping would improve performance.

Thanks again for the feedback! And keep up your good work!


Ross Tucker and Jonathan Dugas said...

To Cowpunk

Thanks for posting again, no problem! I must confess, I went out for a light run after my reply to your earlier comment and the thought was running through my head that Lemond was essentially using this approach to validate claims he's made for a long time.

And part of me is bothered by it, because I know that a lot of people will see his link, come here, and then dismiss it as junk science (as one person did) because they don't read all the posts, or they lack the 'historical context' of having worked up to the final conclusion, when in fact, all the assumptions and potential sources of deviation have been explained.

So yes, this is something of a problem. However, then I get to thinking that this site is, after all, free to view for anyone. So the information is out there - I write the post knowing full well that it may be scrutinized, referred to or that it may be torn down without reading it thoroughly. But it's OK, because the intention is never to provide a definitive answer or the final word. Rather, it's to have the first word and then follow the debate.

And so I post, errors and all, with the confidence that the ultimate conclusion (that the sport is cleaner) is correct.

Re the guys who are tarnished by false accusations, I agree with you, and I'd hate to make blatant allegations that can't be supported. I guess one might say that this kind of analysis should never provide proof, just a line of thought but that ultimately, the process of doping control, and investigation (because doping control has clearly been shown to fall short) will provide proof.

So I appreciate your concerns, and hopefully will have a chance to provide some moderation to allegations without substance!


Anonymous said...

Do you have the data from years past? I would love to see previous climbs, such as the tourmalet and the VAM and watts/kg output?

You mention that is was 6.2 in the past , can you please show specific incidences of this? Like hey lance climbed to la mongie in 2002 in X watts per kilogram.

Anonymous said...

Anonymous said...

Do you have the data from years past? I would love to see previous climbs, such as the tourmalet and the VAM and watts/kg output?

You mention that is was 6.2 in the past , can you please show specific incidences of this? Like hey lance climbed to la mongie in 2002 in X watts per kilogram.

* * * * *

Tour 2002, Stage 11, La Mongie, last 10 km (Distance 10.0 km, Grade 7.9 %, Elevation 790 m)

Time 27:31, Speed 21.80 Kph, VAM 1722 m/h, 6.17 w/kg

Time 27:38, Speed 21.71 Kph, VAM 1715 m/h, 6.13 w/kg

Time 27:44, Speed 21.63 Kph, VAM 1709 m/h, 6.12 w/kg

Tour 2003, Stage 15, Col du Tourmalet, from east (Distance 17.1 km, Grade 7.4 %, Elevation 1268 m)

Armstrong, Ullrich, Mayo, Zubeldia
Time 44:30, Speed 23.06 Kph, VAM 1710 m/h, 6.24 w/kg

Anonymous said...

Just two thoughts:

I think the bio passport has also contributed to the chaos in the sprints, as it is clear that no single team this year was able to control the approach a la the Saeco train for Cipollini or Fassa Bortolo for Petacchi. It was basically an even playing field, and you could see the pattern on wider boulevards, with several sprinting trains forming in parallel (I would place Milram, Cervelo, HTC-Columbia, Garmin, and Lampre, all trying to form a dominant leadaout simultaneously at one point).

Second thought: RBC exchange transfusions have been around since at least the early 1980s (re: Grewal at the LA Olympics). I think I saw one website saying something about Lemond taking EPO, but it wasn't marketed in the first part of his TdF career victories. However, I would think transfusions would be standard practice in the peloton in 1989. His victory on the Champs was one of the most exciting and dramatic finishes in TdF history, but one must also remember it is still the fastest time trial of >20km length. Is there a way to look at the course, conditions, avg speed, and estimate his power output for that half-hour?

fxg said...

I hope I am not the one who views your blog as junk science, because this is far from truth. I wouldn’t spend that much time reading it and all the comments therein if it were the case. Actually, I am quite convinced by your arguments that the low power of this year’s TDF and that of the Giro strongly suggest a cleaner group of riders. However, as a scientist I consider it is a good hypothesis that remains to be backed by more/better data. And I am sure you would agree with that. The problem I noted in my previous post does not come from a “dogmatic” attitude with respect to the 6.2 w/kg. I have never sensed any dogma in your posts. However, you do mention the sentence “physiologically impossible” quite often by referring to your calculations, which in turn make big assumption about efficiency. I’m a scientist in a very different field and your blog motivated me to look for references about efficiency. I found Santella et al. study and considered it rigorous and would have like to hear your thoughts about it. Even more so that they cast serious doubt about the validity of your calculations. I agree that field data do suggest that your ca. 6.2 w/kg value makes sense, but it does not rejects their result. Even if that study has some problems, the wide range of efficiency documented there makes it highly unlikely that a common value can be used to calculate VO2max. This is why the Armstrong case, the only rider for which we have actual data, is interesting to look at objectively. Again, the lower power output does suggest a cleaner TDF, but I don’t think it is possible to rule out that other factors came into play, only time will tell. One major factor that has not been mentioned is the elimination of time-bonuses. I think it removed the insensitive for top contenders to push the pain further to win stages and get those precious time-bonuses. Granted it could not explain the low power up the Tourmalet. Hope my view is clearer. Continue your amazing work!

Joe Garland said...

Re Lemond in the final 1989 TT, early on in this speech, he attempts to debunk how fast it was, pointing to it being largely downhill. In part he is responding to those who claim the TT as proof that he doped.

Ross Tucker and Jonathan Dugas said...

To fxg

Not at all! Your comments are great, they make the site better! That was a couple of other guys in response to previous posts!

And yes, I absolutely agree with the need for more data! Perhaps what we can say is that up to now, we have enough to generate a hypothesis! Now it needs to be tested with years of data, across many riders (not just winners). That, and more research into the Santilla study, which is on the cards!

Thanks for the feedback, as always!


Anonymous said...

This is a serious question: why are you sure Horner is not doping?

Anonymous said...

To anonymous at 10:20

Who said that Horner doesn't use doping? Haven't read that anywhere.
His performance is physiological possible for "normal" humans, which makes it more likely that he doesn't use doping.
It doesn't mean he isn't using, only makes the chance greater he isn't. If I understood correctly.

Ross Tucker and Jonathan Dugas said...

Hi Anonymous:

Indeed, a very good (and fair) question. Answer - I'm not. He may well be doping, you're quite right.

However, his performances are interesting because as the second anonymous poster has said, they're pretty acceptable for a guy riding at 85% of max, efficiency of 24% and max capacity of 75ml/kg/min. So there's nothing there that jumps out.

However, it is important to note that a performance of 5.5W/kg is not necessarily 'clean' - that rider might have been a 5.2W/kg rider without doping and the only reason they're better is doping. However, in the grand scheme of things, it doesn't change the overall conclusion/hypothesis. Horner's data have given a 'barometer' for the Tour, and if he's doping, and it's still dropping, then the hypothesis is even stronger.


Sigmund1 said...

Thank you for your reply to my comments in your former posting. I have read the article you posted last year, and my interest in this subject was somewhat initiated by that post.

Two quick replies to your reply. I certainly agree that fatigue is important within a single stage, my point was more on fatigue over the three weeks. I still think the the GC contenders could produce their best performances towards the end of a three week GT.

Funny that you should yourself express some qualms about having stated a concrete number and not a range. While planning this comment I intended to suggest just that, stating a range instead of a number, I mean.

The problem with a number is, as I think this debate has shown, that people latch on to the number (as I did myself), it gets anchored and we end up discussing how that number appeared and not the concept.

Anyway, you have now made your position abundantly clear so there should be no more room for misunderstanding.

I certainly agree with your assessment of the current situation. The climbing times, and the associated wattage figures strongly suggests that this was a pretty clean race. I think there can be no doubt that Schleck went all out on the Tourmalet. Contador may have had a little bit left in the tank, but not much. If he'd been off his limit he would simply have gone faster and tried to dislodge Schleck more than once.

I am not so sure as to whether Schleck and Contador in a historical perspective are very good. I think possibly not, but thanks to the doping of the last 20 years we shall never know.

And this is one of the saddest things about cycling's recent history. We have been denied a gradual development of the sport so that we have less data on what is physiologically possible given the improvements in equipment, nutrition, training methods and tactics and a deeper talent pool.

I would for example hypothesize that the improvements in pedaling systems (clipless pedals, stiffer soles) better bearings/bottom brackets, better gearing should improve efficiency noticeably over Merckx' days. Also, if the leaders now compete with powermeters, this should improve average power and climbing times due to the improved pacing this allows. Do you know whether there has been done any work on the effects of technology's impact?

Also, in Norway, there have been quite a lot of talk about the training methods utilized by many continental teams to be far from optimal. Our recent history with CC skiing and a very scientific approach to endurance training has produced great results, and many coaches and trainers here; feel that these method have not been adopted in continental cycling. I have been fortunate enough to discuss this with a former pro and he was pretty clear that he felt that the continental training methods around 2000 was very conservative to say the least (at least the ones he was exposed to).

Incidently, this may explain why cycling's nexus of power have shifted away from Italy and France in the last 15 - 20 years.

Anonymous said...

Do you have an opinion on the apparent lower performance of Contador in this years tour compared to last year? Specifically the time trial and also on the climbs? How about the fact that last year Armstrong finished third, (though he was considerably slower this year)? Could the tour have cleaned up in just one year? Is it possible Armstrong was still doping in 2009 despite the risk to his reputation? If he was not is that because he was (close to) the equal of the best?

I wonder also if maybe we aren't all hoping the tour has cleaned up when maybe this isn't entirely the case.

Jonas said...

Anonymous said

"Do you have an opinion on the apparent lower performance of Contador in this years tour compared to last year? Specifically the time trial and also on the climbs? How about the fact that last year Armstrong finished third, (though he was considerably slower this year)? Could the tour have cleaned up in just one year? Is it possible Armstrong was still doping in 2009 despite the risk to his reputation? If he was not is that because he was (close to) the equal of the best?

I wonder also if maybe we aren't all hoping the tour has cleaned up when maybe this isn't entirely the case."
Interesting perspective. I have thought about this myself.

The demise of Lance Armstrong was brutal. I know it's highly speculative, but in my mind, the huge discrepancy between his 2010 performance and 2009 indicates that he was up to something last year. One could wonder whether the Floyd Landis accusations in the Wall Str Journal, or something else in the days leading up to the tour, somehow made it more difficult for him to go through with his routine this year? After all, he sounded real confident in the weeks before, even claiming that he was close to his personal best in one of the Alpine climbs.

As I remember it, the one performance from last year that really stood out was his breakaway at Verbier. Schleck tried to follow and left the other contenders behind, but kept losing time to Contador all the way to the top. If I remeber correctly, Ross/Jonathan estimated that his VAM was in 1870 - 1900 territory (even topping Riis' at Hautacam 96) , which gave a wattage of > 6.5 W/kg for that ascent. I'm not sure what that means exactly, but taken at face value it would indicate that his form in this year's tour on the climbs was significantly poorer. Alternatively, could it be that something was wrong with the Verbier calculations? After all, Schleck only lost about 40 seconds to Contador that day, which means that his wattage was well above 6.2 W/kg (if Contador's numbers are to be believed).

So did Schleck and Contador both perform worse in this year's tour? Or was something off with the Verbier numbers from last year? After all, I seem to remember that performance on the other climbs in 09 were more on par with what was observed this year. The Verbier numbers also beggars belief compared to historical performances from known dopers like Riis and Pantani.

These speculations are not meant to be annoying. I realize that there are no real good answers to these questions...


Phil said...

Perhaps this has been mentioned before (it's been a long series of comments!) but it seems there's a good argument for requiring riders in the Tour to make their SRM (or equivalent) data public. Apart from flagging suspiciously good performances, it would add something to the TV coverage. And if everyone was required to publish the power numbers, there'd be no issue of an unfair competitive advantage.

Sigmund1 said...

Re Jonas, Armstrongs demise this year is easily explained by his numerous crashes.

Up untill this, his performance was quite impressive. He was the best of the GC contenders in the prologue and untill his untimely flat on the cobbled stage he was together with Schleck and Contador.

Contador's most impressive feat last year wasn't Verbier, though that was also quite a performance, but his TT. Beating Cancellara and every other TT specialist and winning the stage was quite spectacular. Even more so when we compare to this year's rather lack lustre performances against the clock.

Ross Tucker and Jonathan Dugas said...

Hi Anonymous and Jonas:

A couple of thoughts of my own.

First, I hope I haven't said anywhere that we're seeing a clean Tour, because I don't think that's true. I think we are seeing a CLEANER Tour, but in any sport, there is bound to be some level of doping, be it cycling, athletics, baseball, tennis, you name it.

So I would be very cautious to suggest that the Tour has cleaned up. I think there is probably "micro-doping", a suppressed level of doping because the biological passport now makes it much more difficult to get away with the same as before. That is, in my view, a significant step forward, and I know that the experts on that bio-passport system are working hard at tightening all the little loopholes that still allow any level of doping.

So we're taking steps in the right direction, but we haven't arrived, and that was most definitely not the intention of my post, and I have to clarify that.

Re the level of Contador and Armstrong this year. Armstrong's decline could well be down to off-the-bike issues around Landis and the Federal case. They could, equally, be explained simply by a crash-riddled stage and a rider who lost desire after that. There are too many "what ifs". However, like Jonas, I am surprised by the magnitude of his fall from that level that we expected, given his pre-Tour form.

So something was amiss - was it doping, was it pressure, was it motivation, was it simply crashing? We don't know. I do think that the Tour could make big strides in one year, and so I wouldn't rule out a drop as a result of testing and better scrutiny. But equally, I can't say this with any certainty either.

As for Contador, his performance on the Verbier last year was good, but it was wind-assisted (which makes the VAM method over-estimate power output) and also, knowing what we do now, the climb was short enough that the performance is not mind-blowingly fast. A following wind (which a number of people have said was the case), combined with a climb of only 20 minutes could, in theory, contribute to a calculated power output of 6.5W/kg, when in fact, it's 6.1W/kg or 6.2W/kg.

This year, on the first part of the Tourmalet, the VAM method was also overestimating slightly, probably because of a following wind. The error was about 0.2W/kg, so that means the value for Contador might well have been shown to be artificially high thanks to this error had we been able to see SRM data for a top GC rider.

But more to the issue of his form, last year, he was definitely better. Now this does happen - there is variability in performance from one year to the next, in any sport. Getting the specific preparation right is difficult - you get sick, you crash, you are forced to rest, and you come back better one time, worse the next! It's not an exact science. So when a guy is slightly off, there's a chance it's easily explained, and I think Contador was not too far worse.

The big drop-off for him came in the time-trial. There's been a lot of talk about how good Andy Schleck was - I'm not convinced that he was, given his position (again this is tough to say because of that wind discrepancy between early and late). But when you look at Menchov putting 90 seconds and 2 minutes on Schleck and COntador respectively, and Sanchez almost matching them, then it was more that Contador came down than Schleck going up.

So that's important, because like Sigmund says, that was his best performance last year. Fatigue? Just a really bad day? A sign of cleaner riding? Who knows? One rider doesn't tell us that, even one Tour doesn't answer it conclusively. It will be interesting to see next year's comparisons.


Frans Rutten said...

Frédéric Portoleau calculated this evening at www.cyclismag.com about 480W(70kg+8kg=78kg)for Andy Schleck's acceleration at the Port de Bales.

Portoleau assumed Andy lost about 50s, IMO too much. Including loss of speed with the restart he lost at the most 35s. The bad luck video(ASO), however, isn't accessible anymore.

But 480W or less for a maximal surge of 4:48 or slighly more.

It's a clear sign, that compared to the past, Andy Schleck or Alberto Contador are significantly slower over 40min. plus climbs during long stages.

Jonas said...

Frans Rutten

Yeps, he lost about 35 secs. The whole thing is here: http://www.youtube.com/watch?v=oe7wpDjGHXU


Unknown said...

Re Ross & Jonathan:
"Armstrong's decline could well be down to off-the-bike issues around Landis and the Federal case. They could, equally, be explained simply by a crash-riddled stage and a rider who lost desire after that. There are too many "what ifs". However, like Jonas, I am surprised by the magnitude of his fall from that level that we expected, given his pre-Tour form.
My thought on Lance's performance after he fell from contention was (and this is sheer speculation), was that he wanted to try for a stage win and that as being in contention for GC pretty much dooms any breakaway from succeeding. So he contented himself for working for Levi to some extent, and took it easy each day after dropping off so that he would no longer be a threat to GC and would later be allowed to go in a break. He did work enough to preserve the team GC position on some stages, but I'm not sure he was even one of the team's top 3 riders for a few stages in the middle. I haven't gone back and analyzed this, but that was my impression as the tour progressed. That would explain his low position on GC time in the end. I may go back and take a look at his GC position but I am pretty sure he allowed it to drop for a while before the first Tourmalet stage.

Anonymous said...

How do you feel about the fact that 26 distance runners at the 2006 European Championships had a hct >50%?


Frank Day said...

I would simply like to point out the big "unknown" in all this theorizing is the issue of efficiency. You wrote: "However, his performances are interesting because as the second anonymous poster has said, they're pretty acceptable for a guy riding at 85% of max, efficiency of 24% and max capacity of 75ml/kg/min. So there's nothing there that jumps out."

None of us, AFAIK, knows what any of these riders efficiency are. So, 6.2 w/kg may be extremely non-physiological at an efficiency of 20-22% and underachieving at an efficiency of 26%. Add on top of that the variation in VO2max and we can see the reason that it is not possible to draw a line in the sand and say above this means doping and below it means clean.

Gene said...

@Anonymous re hct% >50,

If what Ferrari writes is true, raised hematocrit is nothing unexpected for athletes: http://www.53x12.com/do/show?page=article&id=24