Tour de France 2015 - stage 16
  • Tom Wells

    If Team Sky are providing data to UKADA then I don’t really see why there are so many issues around this. I understand people want to see the proof right before their very eyes but Sky don’t want to give away any of their ‘marginal gains’ which I also support.

    I understand people’s skepticism surrounding Sky’s performance so far, but at least put things into perspective. Quintana wasn’t all that far behind and Froome hasn’t been nearly as dominant since either. A good day in the saddle is a good day in the saddle and it looks like that was his day.

    • Burt Fleming

      Tom, numbers are not ‘marginal gains’ they are everything. I wholeheartedly agree with Sky releasing Porte and Thomas’ data as well as talking with Quintana’s team to compare his numbers as well.

      I don’t know what race you were watching to think that Quintana wasn’t far behind mate! When did Porte become a climber? Froome has contested EVERY move so no one can make up ANY time at all. Froome’s dominance is real, whether it be natural or assisted.

      I also can make a case for Froome clean and dirty here. It is not that hard to do. If something is amiss here, it will be in the totality of ‘marginal lies’. I hope Froome is clean, I do, but this SKY side-stepping is getting old.

      • Tom Wells

        A minute isn’t far behind, if we were talking 3 or 4 then maybe and it was after the first rest day as well. It all depends on the situation and how the riders’ legs are feeling. I agree they should release Porte’s and Thomas’ data for that climb too, to be comparable. However, your assertion that other teams should release data to compare to Froome’s is frankly idiotic, why should they? Froome is dominating for sure, but he’s more hanging on now than riding away from everyone. Quintana in yesterday’s stage looked strong, as did Nibali and Contador (even though the latter had a crash, he made up a lot of time on the climb), do you suggest they’re doping too?

        The problem with Froome and Sky is that everyone assumes they’re doping and no-one else is. Which is bollocks.

        I don’t know what world you live in mate, but Porte’s been a climber since he joined the pro peloton…

        • Blake

          “The problem with Froome and Sky is that everyone assumes they’re doping and no-one else is.” Um, no, it’s just that in addition to wearing the MJ, they’ve acted like the shadiest bunch of “anti doping advocates” I’ve ever seen. At this point it’s obvious that Valverde is doping, the question is how can sky beat him so handily and why are they so sketchy about it?

      • pablo_skils

        Burt, mate, I take it you’re not a cycling fan. “When did Porte become a climber?” Er, dude, when did the Pope become a Catholic.

    • Blake

      First of all, I wanted to believe in Sky starting the tour, and I’m still around 50/50 on whether or not they’re actually doping. But Sky is acting really shady so far, and it makes it hard to ignore the signs.

      Start with the leaked data file/video, and the full DEFCON response to 2 year old information (which would be useless or worse to a competitor in 2015). Not only did their lawyers get the video taken down from multiple streaming sites in just a few hours, but they intimidated Twitter (or the user) into closing the original account that tweeted the link. Sounds like some stories about Lance, but applied to a trivial issue… Unless they think the file shows evidence of doping. The response says more than the file ever could.

      On to this years data. If Sky had said “We won’t release that data because it’s a competitive advantage.” I would be OK. Sure, I’d wonder why Pinot, Boom, etc could release their files but sky could not release any info, but it would be a reasonable, plausible position in line with their stated obsession with details.

      Instead they claim 5.78 W/kg, while 2 other riders finished well behind while posting ~5.9W/kg (in actual power files, not a press conference summary). This simply insults our intelligence, and implies the laws of physics don’t apply to Froome. Maybe that’s it, they’re involved in physics doping.

      Sky acts like a team hiding something, that’s why everyone is looking at them.

      • Tom Wells

        Wouldn’t you want a video removed if it even implied one of your riders was doping whether it be true or false? It’s all about the image with Sky and although they were a bit heavy-handed, I can see why they didn’t want it online. Also partly because the numbers don’t actually add up. Why would he be doing more watts in the group than when he’s on the attack? Makes no sense.

        Sky’s numbers released on Monday don’t make a lot of sense either, but that doesn’t imply they’re being ‘shady’. It could have just been a simple mistake, especially regarding Chris’ weight as it can fluctuate a lot based on perspiration among other things. I lose 4 lb on a ride for example, but quickly put it back on after some food and drink.

        And to reply to your post below, I don’t think Valverde is doping myself. Of course this is all just my opinion but this appears to be a clean tour and I’ve watched every stage live so far. CF had a great stage last week but seems to be suffering for it a bit more the last 2 days, struggling to keep in contact with Quintana etc. I can see him losing a bit of time today and tomorrow.

        • Blake

          Would I want it removed? Of course. But there are and will always be people implying things on the interwebs, and treating them like a threat to national security is untenable. Their response more than crossed the line from legitimate concern to cover up. Whether it’s bad PR or hiding doping I don’t know, but if Sky was as good as they think, they would have known to just let it go and die down. Just claim it was fake data for god sakes.

          A simple mistake is what I want to believe, but the most “professional” team in the world, who hangs it’s hat on power analysis and details made a “mistake” that was easily caught by a cat 4 with 2 years experience using a power meter (me)? The only interpretation that makes sense is they intentionally used Froome’s highest weight value and the greatest possible chainring adjustment to present an artificially low number. Not a lie, just an untruth. Which makes me think they want to distract sky fans from the criticisms instead of informing. Agree with weight fluctuation as I experience the same, but Sky knows how to control for that better than we do!

          As to Valverde, I suppose there’s room for belief, but given his attitude during/after puerto, the persistent rumors that microdosing and other methods are undetectable, combined with the fact that he’s as strong (or stronger) as before his ban, I have a hard time giving the benefit of the doubt. It was interesting to see Sky crack today, but it seemed like rivals weren’t interested in hitting him hard. Not sure why Valverde, Contador and Quintana don’t just make a gentleman’s agreement and attack until Froome can’t cover, but I guess they are more concerned with the podium than what step they have. Couldn’t believe Movistar chased Contador yesterday.

  • JasonM

    “I would have made sure that I shore up my argument with all these supporting lines. It is just basic science”

    Gee that’s funny coming from a guy who is basing his argument on estimates and suspicion.

    • Daniel Andrews

      The estimates you just took a backhand to are well grounded in science. All the data from other riders seem to work out in our estimates when we match it with their power meter data. As far as suspicion, it’s hard not to be suspicious when you realize for the Sky data to be true and everyone else’s to be true that Sky must able break to laws of physics. Since the last part cannot be true and the estimates work with every other riders power meter data then Sky’s data must have been massaged, particularly the mass of Froome and correction estimate of Kerrison on their power meters. Gesink, Yates, Ten Dam, and others have their data posted on Strava within 12 hours and the models we use match theirs within 0.5% (one half percent) while it takes Team Sky 7 days to release their data and their estimate of Froome’s power sits between Gesink (1:33 down) and Yates (over 2 minutes down). Yes, I am going to be skeptical of Sky and so should anyone else who passed a secondary school physics class. I am not saying they are doping, but what are they really trying to hide?

      • JasonM

        “what are they really trying to hide?”

        Probably everything. They have a clear competitive advantage. Sitting on the top of the heap why would you extend a hand up to the people you are trying to keep down.

        • Daniel Andrews

          Sky released the cadence data, none of us care about the cadence. All we wanted was average power for the climb and not their interpretation of the data. All the other teams know roughly what the average power was as do we as well from the Strava files and from the models we use. There is absolutely no advantage to knowing power data or heart rate data. In fact the heart rate data means little to us without more longitudinal heart rate data with power data. What would have helped immensely was a body mass that was somewhat accurate with a little fortitude in the response and not “around 67.5 kilos”. Sky’s problem is it can’t hide from the physics involved. The big problem is everyone of us “psuedo-scientists” came to the same conclusion after the Sky presser, how can Froome climb 1 minute and 33 seconds faster than Gesink producing less relative power because it’s physically impossible? That’s an over 10s per kilometer difference from when Froome seperated himself from Gesink on the climb, because they are riding at the same relative power when they were together.

      • PsiSquared

        To have scientific merit, your values should have an uncertainty value, and somewhere you should detail how you account for that uncertainty. Anybody coming out of a lab or doing simulations knows this. Until the people doing these estimations of power output start listing uncertainty values, I won’t take them seriously because it implies one of two things: all of the necessary work wasn’t done, or the uncertainty in the calculation was so large as to cast doubt on your estimation.

        • Daniel Andrews

          Do your own research then. Also, we don’t have to have uncertainty measurements. The Strava files, using power-meters which are an indirect measurements too, have given us the ability validate both the CPL model and the Dr. Ferrari model with 100% certainty within 3%(+/- 1.5%) among the sample files we can measure. Is the sample large enough to be considered airtight? No, it would likely require more than 1000 points of data for climbs from power files taken against 1000 timed rides by video. The only thing we have to do is make assumptions and make those assumptions known. Furthermore, the models and assumptions made by those models and from those models can tolerate a variance range of at least 3% and possibly up to 5%. We don’t claim you can say someone is doping using these models, although if we started getting 6.4 W/kg for riders for 35 minute climbs we could say it is highly likely such a rider is cheating. The best solution would be using RFID chips or NFC beacons to measure power and speed on climbs.

          • PsiSquared

            Interestingly you folks don’t mention any of that when you trot out numbers for the press. And many of you folks do indeed use your calculations to hint that someone is or isn’t doping. Anybody with a science background should doubt your numbers since you don’t talk about uncertainty. It’s basic to scientific studies. Of course most of the public knows nothing of that, so it’s easy to convince them of how good your numbers are. As for claims about the accuracy of your estimations, it takes more than just your claim that they’re accurate. You should at least provide a paper, if not a peer reviewed paper, i.e. the gold standard.

            • Daniel Andrews

              Performance modeling which this is, is not a scientific study or even scientific research. It relies on known science to help form the models. If someone wants to come along and test the models independently and put that out there that would be fine by us. In this case, the known science is physical work and power along with all the physiological correlations we get from such. There is scientific literature from 1930s starting with AV Hill who found a correlation between physical work and power and physiological power. Later in the 1950-1970s, there was Bengt Saltin, Per-Olof Astrand, and Irma Ryhming who were some of the first to normalize power in battery tests to predict VO2max. In the late 1970s Bannister came up with a fitness-fatigue model and from there we have known we have can model performance with reasonable certainty. There is a lot of work by veloclinic and he makes a few references in his work in the development of pVAM and dpVAM residuals model which has shown that we are little more likely to have a grand tour winner who is clean than compared to the previous doping era. People like myself and others analyze outputs of CPL and Dr. Ferrari models which are based off the work of Drs. Michele Ferrari and Franseco Conconi of known blood doping fame, but nonetheless their work has stood the test of time. So there is plenty of peer-review papers out there which support what we are doing from about any which way you want to slice and dissect what we are doing. You seem like a reasonable person and not one to gish gallop on us, but there is no way we can go about addressing

              We try to get the best data possible we can use. That doesn’t mean we really need an exact body mass, but it helps when we want use the power data from strava. However having 1 exact body mass (within .2 kg of what it really is) would allow us to estimate everyone else’s body mass with the same certainty because the relationship is dependent on body mass and relative power at least on climbs relates to speed or time more precisely. We know everyone has to do the same amount of relative physical work with their body mass to ascend the climb. Is it perfect? No, models aren’t perfect they only need to be valid with a certain amount of confidence. We know wind can play a role which makes the CPL a better method to use than Dr. Ferrari method, but this is only in the case a significant tailwind and we rely on interviews of the riders to give us such information. Validity and maintaining confidence is what makes a model sound in its application which is what we are doing while testing said validity and confidence at the same time. This is the major deviation from science in the process as modelling requires a truly stochastic process in the refinement of the models and fits more into the realm of engineering than actual science because science requires a more rigorous and controlled process of refinement. Science does however explain how the models work and why they can be valid. Just as science comes to explain everything we do know.

              We also are not responsible for what other write about our processes. Yes, we understand there is a possibility of compounding error especially in models which use other models taken from indirect measurements. When you go to Veloclinic’s blog, Dr. Tucker’s website, and my blog you see caution, questions, and skepticism, but also a call for transparency and openness.

              You are a smart and reasonable person. You have to know we can’t go chasing all the gish-gallops we are getting from most of the people who say we are using “psuedo-science”. If we were using “psuedo-science” it should be quite easy to for them to disprove what we are doing by using science. I am not going to go ask Sky to prove the power-meter correction is 6%. I am going to make the assumption the manufacturer knows more than Sky. I am not going to assume Froome weighs “about” 67.5 kilos. I am going to use their average power data because that was the only thing they said with any certainty, I am going to work with a range for Froome’s mass, and I am going to work with a range for the correction for the power meter which goes .5% beyond the manufacturers range. Once you put all that into a chart and then match it with the model the most likely values for correction on the power-meter is between 4.0-4.7% and Froome likely weighing between 65-66.6 kilos which gives us 5.96-6.11 w/kg. We have no reason to believe Froome has a mass of less than 65 kilos, but we can almost claim with certainty he didn’t have a mass of 67.5 kilos at the end of stage 10. Froome’s avg. absolute power given by Sky doesn’t differ from our estimates significantly, his VAM doesn’t differ from our estimates. So his relative power can’t differ from those by a greater percentage. https://uploads.disquscdn.com/images/d08a3333a9636c71559e2d57fafcf9b3d9d9dc794d23721e4a02dd6fbf5e53db.png

          • Bart Scicchitano

            How do you know the weight of Gesink, Ten Dam etc? If it is based on what their strava profile says, then how do you know that is accurate (how often do you think they bother to edit their profile)?

            And if you must insist on trying to discredit every winning performance in World Tour cycling, can you please start asking these same questions of Nibali and Contador?

            • Daniel Andrews

              here are my thoughts from last years tour.

              http://sprenten.tumblr.com/post/92853257596/skeptical-about-top-cycling-performances-in-grand

              Lastly, no one is discrediting anything. We asked for data and not interpretation of said data. We can interpret the data just fine which is why we are very critical of their interpretation of their own data. As for their reasoning in botching it up so bad. I haven’t a clue. They only created more questions.

              • Bart Scicchitano

                1. Why are you the one entitled to the data for a “fair and unbiased” analysis? You are disgruntled because you asked for something and didn’t immediately get what you wanted?

                They have been passing data on to anti doping bodies. These are the people charged with identifying “superhuman” performances. The information passed on at the press conference was going beyond reasonable expectations of any cycling team (no one is asking these question of Astana, Saxo etc), but for some reason you hold Sky to a different standard?

                2. You are on a pedestal proclaiming something is awry because Sky didn’t give you the exact numbers you demanded of them – Should I jump on my pedestal now and proclaim your work is tainted because you failed to adress my question of the weights you used to calculate watts per kilo for the Gesink, Ten Dam etc data?

                • Daniel Andrews

                  British Cycling used the same CPL model we have applied to Gesink’s data.

                  http://journals.lww.com/acsm-msse/Fulltext/2011/05001/Critical_Power_And_Aerodynamic_Drag_Accurately.488.aspx

                  Everyone has been using Martin’s, Ferrari’s, and Conconi work.

                  In the end, mass is only needed to validate the models work and mass doesn’t have be precise because we work within ranges with models and we can extrapolate to near what a riders true mass is with enough data points. For Gesink, I use a range of 70-72kg. For Froome, I am using a range of 65-67kg after narrowing it from 65-68kg. My OPINION, is Froome is likely between 65.5-66.5 kg at the start of the final climbs and that Gesink is somewhere around 70-71kg.

                  Sky’s power-meter data isn’t bad, just the interpretation they gave of the data was bad. Please separate the two. Gesink has a higher VAM listed by Strava than the interpreted VAM of Froome according Sky, that’s not possible. If Sky gave an average power, distance, and time for the power reading along with a specific mass we wouldn’t be having this conversation. They botched the data analysis or they purposely tried to discredit us by giving a bad interpretation.

                  • Bart Scicchitano

                    As stated in the above article:

                    “He said that the numbers pertained to a 15.3 kilometre section, which he took as the starting point, and which was covered in 41 and a half minutes … a recorded average power of 414 Watts.”

                    There are your distance, time and average power for the period in question. The question of 4% or 6% overestimate is up in the air. The amount by which the values are affected is not going to be a consistent for all riders or chain rings. I would think the value given by Sky are what they measured in their own tests.

                    Weight has also been clearly stated as “around 67.5” and you have said “mass doesn’t have to be precise”. So there is the final piece of your “average power, distance, and time for the power reading along with a specific mass” shopping list.

                    I don’t think the power data is wrong. I think your insistence that the interpretation presented MUST be incorrect is wrong. You say you just needed these specific values, but even when given them, your bias drives you to come up with different values that support your view.

                    That isn’t science. It’s just someone with a pedestal and an agenda.

                    • Daniel Andrews

                      We don’t doubt the recorded Watts from Sky, It is the calculated values of VAM and W/kg are the ones we don’t understand from Sky. Froome needed a larger VAM and larger W/kg to be 1:33 faster than Gesink for a 40 minute ride. I’ll use the Alpe d’ Huez ride between Froome and Quintana as an example as they had a 1:20 gap which is a 13s than the one between Froome and Gesink on St. Martin. The difference between Quintana (6.09, 1712) and Froome (5.89,1656) is 0.20 W/kg while the difference between Froome (6.07,1657) and Gesink (5.85,1597) on St. Martin is 0.22 W/kg using the Dr. F. formula. Gesink would have had to have a mass 73.6kg for the values of Sky to be correct using Gesink’s data from Strava. That’s not standing on a pedestal with an agenda, it’s asking an honest question, either sky calculations are correct and the models or strava data is wrong or the sky calculations are incorrect and models and strava data are correct. It has to be one or the other, both can’t be correct or we are breaking the known laws of physics.

                      Yes, I find Froome’s ride on St. Martin to be closer to the edge of reason than I would like, but as myself and others always point out it is only 1 pixel in a larger picture. Froome actually shows what we would expect of a clean riding peaking during and fading slightly over the course of the Tour. While on the other hand we see Quintana actually getting stronger and the 1712 VAM on Le Alpe raises more questions than Froome’s St. Martin climb. You can’t have it both ways, all the initial indications of the all work we have done point to the peloton as a whole is likely becoming cleaner this is the good news. According to what we know about human performance all the top performances at the Tour point at least fit into the gray area. Quintana improving like Nibala last year on the last climb is reminiscent of the doping era. That’s something which is worrying. We are still getting performances which equal or surpass those of the doping era. This is also worrying and the benchmark for this is Valverde who is performing as good if not better than he ever has. In Quintana’s defense it is possible to improve at a Grand Tour as a clean rider, but improving to the point he did this year still raises questions as it’s not exactly like Nibali who had two rides last year which raised eyebrows. We don’t have a bias in the way you think, we are skeptical and being skeptical is healthy as opposed to cynical.

                      The one who does have a bias is the one who can’t think that Sky is possibly wrong in their calculations. I already pointed out that our models and strava’s data maybe incorrect if indeed Sky is correct. You have not and are the one with an axe to grind.

  • Jonty Paulsen

    Well done Shane, Another Great article you along with Paul Kimmage are probably in my opinion the two most respected, honest, and knowledgeable cycling journalist in the game today.. You don’t let the BS get into your journalism you speak the truth and have no bias.

    This pretty much sums up the whole article and Team Sky:
    “You have Froome at 5.78 watts per kilo, gaining two minutes on a guy [Yates] who is at 5.84. One and a half minutes at a guy who is at 5.9 [Gesink] and four and a half minutes on a guy who is 5.5 [Ten Dam].

    If they numbers don’t add up then they don’t add up pretty simple!

    • Dave

      It means nothing if they are not all using the same type of power meter.

      • Jonty Paulsen

        Yeah okay, That’s the crap sky and their bias control and manipulation of the media will feed you. So you’re saying It also means nothing when i go up the hill at 24km/h but my mate Alberto Jones climbs it at 22km/h and he stills beats me..

        • Dave

          I’m no fan of Sky, but they urgently need to fire the PR person who thought feeding the trolls would be the best idea.

          It shows the problems of the internet quite clearly, in the old days this sort of article would have been printed on paper and therefore good for at least something even if it was just for wrapping chips or starting a fire.

          • Jonty Paulsen

            All I want is the truth whether sky are doping or not I mean it’s pro cycling, Will it ever be clean? NO. But Sky racing chose to treat all of us cycling fans ( At least With a half decent knowledge not just that couch slobs who want the tour and that’s it once a year ) with complete disrespect releasing such false and manipulative figures trying to show that their a “clean” team while almost in a way now trying to point the finger at other teams for not releasing their figures. Also what’s with the way they choose within the team which convicted dopers are bad guys and which guys are the good guys…

            • Dave

              “All I want is the truth whether sky are doping or not …”
              Then why are you wasting your time on an article about power data? You won’t find the answer here either way.

              “… I mean it’s pro cycling, Will it ever be clean? NO.”

              Very much correct. But I’m failing to see any evidence that Sky is any cleaner or dirtier than the acceptable level of doping (when is dope not dope? when it’s a “supplement”) in all the other teams around them.

          • Daniel Andrews

            They don’t have to be using the same power meter (we don’t really need the power meters with many thanks to the Italian doping doctors of the 70s and 80s), they really only need to be doing the same climb. What is measured is actual physical work against gravity (m * d). To get power you divide by time and to normalize the data you divide by body weight. Since they all climbed the same climb on the same day at the same time and there were many attacks we can throw out most of problems of drafting which is negligible going uphill and wind. What the power meters would be useful for is measuring mechanical efficiency of the bike and pedal stroke efficiency of a rider. Which still doesn’t change the requirement of the physical work which needs to be done.

          • Daniel Andrews

            They don’t have to be using the same power meter (we don’t really need the power meters with many thanks to the Italian doping doctors of the 70s and 80s), they really only need to be doing the same climb. What is measured is actual physical work against gravity (m * d). To get power you divide by time and to normalize the data you divide by body weight. Since they all climbed the same climb on the same day at the same time and there were many attacks we can throw out most of problems of drafting which is negligible going uphill and wind. What the power meters would be useful for is measuring mechanical efficiency of the bike and pedal stroke efficiency of a rider. Which still doesn’t change the requirement of the physical work which needs to be done.

        • Cam

          A power to weight comparison is not a direct comparison with speed. Unless the riders are exactly the same weight and have the same aerodynamic profile you are comparing apples with oranges looking just at power to weight.

          Although a 15kg heavier rider riding up the same hill at the same speed will require a lot more power, the power to weight ratio will not need to be as high, the lower the gradient the bigger the difference in power to weight requirement.

          Yes comparisons can be made, but to say 5.6w/kg cannot go faster than 5.8w/kg is completely false.

          • Daniel Andrews

            It’s called normalization and it does have a direct correlation with speed with respect to climbing. The power-meter on a bike really is measuring torque (angular) which is then converted to watts (linear) which is what we are measuring and is just as much an indirect measurement of work and power as the estimates every one loves to hate. However you want to slice it, moving 100 kilos up 1km over a 15km or moving 120 kilos up and over the same distances at the same speed will require the same output of relative power. Aerodynamics on a bike going uphill aren’t worth putting in the equation as the difference between two riders is going to be less 1% (+/- .5%)

        • PsiSquared

          Nope. It’s a documented fact that power meters from different manufacturers generally don’t read the same. Hell, it’s likely that power meters from the same manufacturer don’t read the same. As for you and your mates, the speed only matters when it’s accompanied by power numbers and weights, and by weights I mean total weights. More importantly, any real, credible analysis would either do a test to compare the power meters and determine how their values differ in a controlled test or said analysis would state that the power meters weren’t tested together and as such, some uncertainty accompanies the comparison of power values.

    • Eat More Lard

      I find this article disappointing, well below what I expect from Cycling Tips. It seems to be a platform for Mr Tucker. He says 5.78W/kg, how can that be right, does a few calcs and then arrives at 6.02W/kg, saying that now it starts to add up. Correct if I am wrong, but that’s a 4% difference, well within the margins of error or variations in power meter calibration etc. If you are going to sling the mud, you kind of need to start from a solid foundation, not one made of sand. I’m no Sky fanboy, but I find the constant accusations beyond tedious. They’ve got to the point where I now don’t believe anything I read in relation to “evidence of superhuman performance”, or worse the passive aggressive “well it doesn’t prove doping but” argument. Ironically, I find myself in the opposite place than during the dark ages where it was hard to believe anything in relation to a clean performance!

      • Nemeseri Andras

        Couldn’t agree anymore.

        Even without the Osymetric chainrings 4% is nothing in the current world of power meters. Power meters have +/-2% accuracy in general. Stages is a left only power meter and doubles the power transferred to the left crank arm. This might lead to additional accuracy issues in certain cases.

        But even if they have used the same pioneer power meter, it would have been difficult to compare data just because power meter calibration. Stages has auto calibration, but it certainly works differently than Gesink’s pioneer that might have been calibrated 2 days ago at a different temperature at a different altitude. It’s even hard to compare two different bike equipped with the same power meter with the same rider on the same course.

        But you don’t even have to look into the power meter accuracy to make it impossible to compare different rider’s data. Athletes can loose weight during a stage easily. Also you can’t be sure that they have the same additional weight on them, like bikes, water bottles, shoes, helmet, clothing, transmitters etc. These things can add up easily.

        Comparing rider power output and w/kg is only possible in a laboratory with proper testing protocols. I’m sad that this article suggests otherwise.

        • Daniel Andrews

          Each and every climb has a workload requirement which when normalized to body mass is the same for every rider. It’s physics, it takes the same relative workload for you, me, Froome, or anyone else to climb x number of meters, if we do complete climb in the same time we have the same relative power. If you complete the climb faster than myself, your relative power will be higher than mine. Simple and secondary school physics problem this is.

          • Nemeseri Andras

            The only problem is that you use different devices with different calibration to measure each riders’ power. Also you just estimate rider weight. That gives you a huge margin of error. Enough to make the whole comparison meaningless.

            • Daniel Andrews

              We only use the power-meter to test our models validity. Models don’t have to have precision and can’t have precision if they are going to be fault tolerant, this allows the use of ranges. The question is what range is acceptable? If a model input has a wide range of known error we wouldn’t use it, but lets take for instance the correction factor for Froome’s power meter, the range of error for the correction is 1% by the manufacturers reccomendation, but by creating a chart and using a range of mass for Gesink to test against as Froome has to be within a certain W/kg range to climb faster than Gesink we find that the model can stand a correction range from 3.5%-5.5% within the limits of what Gesink’s and Froome’s could possibly be. For Sky to be correct using those values for error correction on the power meter Gesink can be anywhere from 70-72Kg if Froome weighs 67kg which is .5kg lower than the value Sky gave for Froome’s mass. If Froome did weigh 67.5kg then Gesink has to weigh 73kg. The second problem is the speed of the climb, because VAM and W/kg no longer start to correlate with each other and the only way for that to happen is be wrong about the incline.

              What the rider’s mass is inconsequential when you have enough data points because the relative demands of a climb will allow you figure out each riders mass to a range through regression with a degree of certainty. If precision mattered then the whole comparison would be meaningless, but model validity is based on not being able to create a situation which defies known knowledge in this case it is the actual physical work being done and the related power to do this over a specific period of time. This is why we came out guns firing with Sky’s interpretation of their data, and not against the actual data itself, just the assumptions made by Sky which made their data not to fit any model creating a situation which can’t exist in reality (my belief was Sky wanted to create the illusion the model is not valid).

              Keep in mind the systems engineering aspects of modeling. The modeling process doesn’t do anything but allow us to gain some insight in terms of performance in the absence of direct measurement techniques, predict future performance with a degree of certainty, test the model against other measurement techniques, and when a known model is known to be valid it can used to take apart someone else’s misinterpretation of data to find the reason for error (reverse engineering). More on this to come later.

              The link below is a quality peer-reviewed article on modeling with respect to cycling.

              https://www.researchgate.net/profile/Tim_Olds/publication/11911435_Modelling_human_locomotion_applications_to_cycling/links/02e7e52f5a49627f86000000.pdf

  • PetitTour

    I do value Shane Stokes as a journalist, but I feel these articles about Chris Froome’s performances come across as a platform for Ross Tucker more than anything else. There are plenty of credible sports science experts out there who have different views to Tucker, yet Stokes chooses to give air time to Tucker. All this provides a rather narrow, one-sided, view of the debate.

    • Garaham Arnold

      Yeah those “sport scientists” Are either employed by SKY, Or of British nationality who control most insights of what we see and hear in Western media.

  • Michael Phan

    For any given speed:
    Low power = high cadence.
    High power = low cadence.
    Simple…

    • Dave

      You’re talking about torque, not power.

  • Kells

    All the talk about power, climbing rates, etc is guesswork and in a lot of cases just pseudo-science. A good day in the saddle is not proof of doping and all that the likes of Kimmage, Vayer and Stokes are doing is asking Sky to prove a negative which of course they can’t.

    When Lance Armstrong was doping, David Walsh had solid evidence from six people within 3 years who had been in his camp. That’s just one journalist. There are always going to be disgruntled employee and leaks in something this high profile. So far there has not even been the slightest whisper to even one journalist anywhere about Chris Froome. Either Sky are incredibly good at keeping secrets or it just isn’t happening. Either way it is up to said journalist to actually go and find evidence.

    • Burt Fleming

      Too many people are stuck on the ‘can’t prove a negative’ train of thought. Asking for numbers, from a self-described transparent team that is suddenly as transparent as a brick wall is not asking them to ‘prove a negative’, it is asking for a team to help prove a positive, a very positive result from Stage 10. Stage 10 wasn’t just a ‘good day in the saddle’, as if!!! LOL!!! How many of those days have you enjoyed Kells?

      The key to all of this is Porte’s numbers compared to Quintana’s. SKY can release real, imagined or fudged numbers, how do you know they are real to begin with? No one knows, for now. The problem is SKY needs to distance themselves from all of this release of numbers/transparency fight because the numbers now are concrete! Once pressure grows for Porte’s numbers and Quintana’s are published, SKY will have no way to back out of this first report. Careful what you wish for…

      Also, it took YEARS and a determined Walsh to collect his evidence. If something is wrong in the SKY camp it will be revealed later than sooner. Walsh suffered going after a ‘clean’, ‘never tested positive’ hero in Lance Armstrong. Walsh had no disgruntled ex-employees, he had to gain the trust of people that were afraid of Armstrong. If history does repeat itself, then this is all so eerily similar to Lance in 1999 and his climb up Sestriere! Just as Walsh doubted Lance, a man he admired, after seeing that climb, so are fans/journalist saying the same about Froome and La Pierre Saint Martin. I like Froome, I’m hoping he’s clean but information like this is starting to sway me the other way.

      • Kells

        It was a good day in the saddle. Even if the number are adjusted giving Froome the minimum weight (66 kg) and reducing the asymmetric ring to the 4% figure the manufacturers gave then that would be a w/kg figure of 6.1. Good, but way way less than the 6.4-6.5 the likes of Armstrong were pulling. Even the sports science website that takes such delight in chasing Froome give a threshold of 6.2 for real suspicion. So no, it was a good ride but not extraordinary. If you think different then I don’t know how you reached your conclusion, perhaps you could tell me.

        As for releasing all their data why should they bother? You just said now how do you know they are real to begin with. It doesn’t matter what Sky do, the trolls will always be out on the likes of France-2 quoting silly rubbish like 7.0 w/kg. A complete joke.

        And Walsh had his Armstrong contacts within 3 years of beginning his investigation, 6 of them – and that was at a time when other journalist didn’t want to know and there was a general atmosphere of non-suspicions. The whole of France seems to be after Froome as well as a bunch of other journalist trying to make a name for themselves. Despite this there has not been a single sniff of real evidence. And that is after the same 3 years of chasing him.

        This is nothing like Lance.

        • Asor.Rosa

          I find you wrong on a number of levels, blinded by admiration and national pride just like David Walsh.

          First, the whole of France is not at all after Froome, actually not even a tiny part of it is… Indeed, there is a deal of sensationalism in the way the media treat these issues (they follow a Murdoch recipe after all, the prevailing model for popular media nowadays that gets to fund cycling teams among other things) and of course this is made worse by the frustration of spending almost three decades without a french winner of the tour, amidst doping scandals concerning primarily the non-French, post-festina dominators of the tour. But they treated Laurent Jalabert just as harshly when his retroactive EPO positive came out and then they subsequently forgot about it, as they will with Froome when he takes the yellow jersey home, as he will, and stop showing it off around France.

          Secondly, even if there is a lot of excess and hysteria within this power/numbers debate, coming from the likes of Vayer for reasons similar to the ones stated above as well as his inside knowledge of the field, you couldn’t be more wrong and blinded in your call to more or less forget about it. I concede in advance any reservation on the comparability of different numbers debated right now, but proposing to keep an eye on performance next to testing urine and blood, and comparing averages from different days, or even different races and different years, to see how the abilities of an athlete evolve seems to me like a reasonable concept, full of potential. Even if it is not entirely operational yet and will never be a direct tool for establishing violations, the study of power outputs (or even estimates and averages, like Vayer’s calculations, proven quite precise and used by himself within the whole of a three week tour to draw ”indications” and not conclusions) is part of how we understand cycling now and you have to deal with it.

          Last but not least, ”this is nothing like Lance” is a funny statement. Lance himself was arguably nothing like before, nothing like the 90’s, nothing like Puerto and nothing like Landis. And yet so similar…Every case has its context, and it is very probable that whatever happens nowadays, if anything, will prove similarly different to Lance, to Festina and even Merckx, yielding different results (6.1/6.2 w/kg rather than 6.5 maybe) and practiced in different ways, thus providing different evidence. But you have to be a real sky hooligan to oversee certain points that raise suspicion. Yes, we haven’t heard someone linked with Team Sky assert such things as Emma O’Reilly or Betsy Andreu etc. but again we are probably talking about different practices, entailing much simpler, more streamlined logistics and a much reduced omerta in terms of the number of people concerned. Brailsford is notorious for his attention to detail, a new-school guy that is in no way arrogant and used to the old ways like Bruyneel. He knows he cannot count on the Omerta as Bruyneel thought he could. But he has showed a suspicious attitude: Leinders was a notorious doping doctor, therefore paid primarily in this goal rather than to treat injuries and illnesses by a number of teams before Sky. Why would he be hired to treat road rash suddenly? TUE’s, Ventolin for imaginary Asthma and Tramadol, also show a dodgy attitude to performance enhancement. They are all signs for suspicion, as were, at the time, O’Reilly’s and Andreu’s assertions before they were proven to be true (which they weren’t by facts, just by LA’s admission).

          We are all swimming in muddy waters with regard to doping, consequence of our sport being way ahead of the lot in its consciousness of the problem. And yet we can still enjoy beautiful races, as I did with today’s amazing spectacle of a stage. By forcing a simplistic narrative you neither help us collectively swim, nor do you increase anyone’s viewing pleasure; you just reassert your own wishful thinking. Hoping I didn’t sound offensive, I tend to think there is no place for this in cycling anymore.

          • Daniel Andrews

            This may be the single best response I have seen in a while to this whole mess. The so-called “psuedo-scientists” can’t be dumped into the same bin as those “anti-Froome” or “anti-Sky” who believe Froome and Sky are doping. In fact, we don’t really care if he is or if he is not, we already know the likely winner and other 2 on the podium will likely be in that suspicious zone of performance of which we can neither prove or disprove without a lot of other variables. All we have ever wanted was greater transparency and Sky’s actions have always been hostile to us in much the same way the dopers of previous generations acted towards their enemies. If anything the “psuedo-scientists” are Sky’s biggest allies if they and their riders are indeed clean.

        • Daniel Andrews

          Just an FYI, Armstrong had maybe 1 climb in the 6.4 range. The climb to Plateau de Beille was done within 1s of both Basso and Armstrong by all the major contenders of this years tour. This didn’t help Sky’s cause or any other riders cause. Pinot power meter average works out to the 5.9-6.0 range. In 2013 Froome was equalling or better Armstrong’s times, in 2014 Nibali was doing the same. Those are all major concerns, but from the work of Dr. Mike (veloclinic) we do know the field is likely cleaner and there is more work to be done. It is possible to win a Grand Tour clean, but there is no certainty the winner is clean as the winner and rest of those on the podium are almost always going to have to be in the zone of suspicion riding 5.9-6.1 W/kg for most of the major tour climbs.

    • Daniel Andrews

      We don’t really care about the doping part of the conversation which is taking place. The only way to a cleaner sport is greater transparency, we believe our models are correct which leaves Froome sitting squarely in the suspicious zone. The problem we have is physics is on our side at least with relation to power data, we just don’t see how Sky’s power data holds any key to a competitive advantage other than your opponents need to know what kind of power they have to hit to win the Tour or any other Grand Tour and the problem with this argument is we already know you a need 5.8 W/kg for a 40-45 minute climb to be in the front group and 6.1 to win a Grand Tour. Those markers were born out of our models you claim are psuedo-science, but they are science and a lot of it simple physics. Yes, we have to make assumptions, but despite this our models work for every other rider so why don’t they work for Team Sky? That’s were we get skeptical and we don’t misinterpret maximum aerobic power for average power either, we know the difference between those terms. No said Team Sky is doping, but their actions with regards to transparency don’t help us think of Froome or Team Sky with any less skepticism it only makes it worse.

  • _kw

    In my humble opinion Sky is just asking for this media witch hunt by belittling other teams with their marginal gains talk (as if these teams still employed an approach of the 70s) and their reasoning regarding the competitive edge with power data does not convince me either.

    An MMP chart only tells you what the maximum power a rider was able to sustain over a certain time within a timeframe. The longer the timeframe, the less you know whether the rider would (i) actually be able to put out that power in an upcoming ride and (ii) be able to reach the values in his MMP chart for several interval lengths during one race, stage or during successive stages of a stage race.

    The magazine Peloton ran an interesting article recently about Thibaut Pinot’s brother publishing a scientific article with Pinot’s power data / power profile and its improvement over several years since he had – at an early stage in his career – invested in his first SRM. Clearly this disclosure and not a DI2 failure and immature reaction to it must clearly be why Pinot is not leading the Tour right now.

    To me, the power data released publicly does not help anything and I agree with both Tucker and Greg LeMond (who also used the first SRM units) that they – as well as other GC contenders – should release more substantial power data. This does not necessarily mean releasing your directory of training files but some verifiable analysis of it.

    Don’t get me wrong, I enjoy following the races and am by no means naive regarding the performances by some athletes but just blaming it on the press and the fact that it will take years to overcome the Armstrong-era is too simple of an explanation to me. I do agree with Froome that certain French former pros are pretty bold in their statements given their past…

  • Derek Maher

    I guess Chris Froome is passing all the tests at the end of the stages the medical control are taking so one assumes there is no problem with him.
    Plus his team riders are also passing the tests as we have had no word about any of them testing positive for illegal substances.
    So its possible to conclude that the Sky Teams rivals are performing very badly in this Tour.
    The media are having fun,The French are annoyed at yet another Tour without a French yellow jersey contender.The cycling bosses are as usual shooting the sport in the foot with non stop talk of doping and the need to drag people from their beds to take intimate samples from their persons.
    At more expense from cycling coffers.Oh well it gives the crusaders fun.

  • Marcus Mendez

    this is all stupid, where exactly did yates get dropped,…getting back to w/kg, 7 w/kg or 6 w/kg , that’s a 16% difference. 5.8 to 6 that is 3.4% difference. Feels like people are getting dumber over time regarding basic arithmetic, without even getting into statistics , error etc . Here’s is what we do, we make them all ride the same bikes, same tires, and we have a wind speed adjuster. Hmmm maybe wecan get them to ride the same cadence and somehow get them to have the same aero profile….it’s like the world is going socialist/communist on cyclist performance.

    • Dave

      As well as middle school maths and basic statistics, you can add in complete ignorance of the scientific method too.

      All this talk about power data is absolutely useless for determining whether Froome could be the first non-doped GC winner since WWII. Without controlling all the variables, all it does is provide a more interesting way for hacks like Stokes to say “holy crap he’s fast.”

  • Burt Fleming

    “Team Sky insists that was unaware of his past when it brought him on board and maintains that he (Geert Leinders) didn’t do anything unethical while with the team.”

    This is without any doubt one of the most laughable and concerning comments of them all!!! I remember Lance talking about how he believed Dr. Ferrari to be a good and ethical man… Wow!

  • powerboy

    Before jumping to any conclusions about the data released by Sky, one should consider the accuracy of the data they provide. Assuming they use Stages like they claim to do!

    DC Rainmaker has some very valid critique against Stages power meters (one-sided vs two-sided power). Essentially the absolute power numbers provided by Stages can be off for a number of reasons, and thus doing any comparisons based on these numbers is a bit silly.

  • Rupert Rivett

    Power output measurements still don’t relate to drug taking or do they? ……… nope they definitely don’t, really don’t know why anybody would think that having the figures would make call for saying Froome has cheated. After all readings can be wrong and 4 to 6 % accuracy is a massive difference at this level in the sport.
    I just think this whole thing is going in the wrong direction, it proves nothing even if the figures could be relied on.

    • Daniel Andrews

      The power measurements themselves can’t confirm doping. What the PO models can do is confirm with certainty when mixed with other data that a cyclist should be tested regularly, often, and randomly with regards to time of day once they start putting up to many values in the suspicious category.

  • drayzen

    The core problem here is that it’s an apples to oranges comparison of the core power data being referenced as they’re all using different power hardware. Even if they were all using the same hardware it would still be an issue. Being a retailer forInfoCrank I’ve had some long discussions with the makers in relation to how it works vs other brands. I know this will sound like a sales pitch but it’s not. I’m just relaying what I was told and it would appear that the bulk of the power products out there can not be directly compared against one another accurately. The key thing I was told was in relation to their supply of units to the UK Sports Institute. They initially supplied them with a few test units, which then increased to 18. They had previously been using 60 SRM units.
    Now this is the key point here as we’re trying to compare data from multiple power units, they were regularly calibrating their SRM units so that they were able to compare the data against one another and thus have accurate and meaningful comparative data for one athlete against another. Without the calibration the data would be inaccurate and thus only a rough approximation.
    Now look at what were talking about here, not just trying to compare uncalibrated SRM units, but different brands with entirely different technologies.
    The makers of InfoCrank also told me that it’s the only power solution that is tested and certified as a measuring device, I haven’t personally checked if any of the other brands have gained equivalent certification, but as I understand it, the way they work by calculating instead of measuring means they wouldn’t be able to. As far as I’m aware all other power solutions are providing figures based upon calculations/estimations/averages while InfoCrank provide direct measurement. As such they don’t require calibration and data from multiple units can be directly compared against one another, which would be key to this discussion.
    I’d expect this would actually be why the calculations done by Ross Tucker appear all over the place, the tools (power meters) being used to collect the data are not directly comparable so neither is the data…

  • Superpilot

    I’m no Sky puppet, insert any other team into yellow (imagine if Nibali was leading, how the race based suspicion would fly over the moon!) and you can see that it doesn’t matter who is leading, and it doesn’t matter how much they try to explain, but they can never do enough. The media and the fans drive this, we all click on the stories and love the controversy. There will never be a tour without this speculation, ever. So sad.

  • lickspittle

    Ross Tucker on where he got Gesink’s weight from in order to make such accurate comparisons to Froome’s data – “Mean of rider self-reported/reported weights”
    In other words he found it somewhere on the internet. May as well have pulled it out of his arse.
    Pseudo-scientist gives this clown too much credit.

    • Daniel Andrews

      You have the Team site and you have the Strava site. Yes we don’t know exactly what we each riders weight is, but using science/math we can come to the conclusion based on everyone else’s self reported weight on the internet and start fitting these values to a line and regress to everyone’s true mass with 99% confidence.

      • lickspittle

        If Brailsford were lying there are 21 other teams who could blow his figures out of the water. Leading to his disgrace and, probably, the disqualification and disbandment of team Sky. You know, the same play that Ross Tucker tried with his made up guesstimates but using real data. There’s absolutely no way Sky would dare to risk this.

        • Daniel Andrews

          Disqualification for what? For fudging up data? If we were lying, we would have been blown out of the water by now and it hasn’t happened. I think the other teams like seeing Team Sky implode especially the ones which allow their riders to publicly post their data.

          • lickspittle

            I don’t even know who you are but Tucker is holed below the water line. He’s had to retract part of his accusations referring to Yates data (which he simplay made up) and he’s taking a pasting all over the place. As for Sallet and his ludicrous 7w/kg claim, do even you believe that he’s credible?
            And yeah, if Sky had been seen to have fudged the data they’d be toast.

            • Daniel Andrews

              Sallet was referring to Maximum Aerobic Power and made a reference to 5.98-6.23 for his relative power which is wide range of which most of us in Dr. Tucker’s camp are in the lower range. However why isn’t Sky taking a pasting for it’s 6% correction when the manufacturer says 4-5%? No Dr. Tucker is not taking a pasting, he just has to defend himself against the gish-gallopers.

              • lickspittle

                Now I know who you are, I checked your site. You’re another person using estimated weights of riders to massage the figures to suit your biases. Weights that you can’t possibly know with any accuracy.
                And again you have no answer as to why Sky would deliver a hostage to fortune to the other teams by fudging the figures. They’d be ripped apart if any of the other teams or even an individual rider came up with figures that undermined those they gave for Froome.

                • Daniel Andrews

                  Yes, the estimates are correct to. You can check them against their Strava files for those who have them online.

                  • lickspittle

                    Again…
                    Which data? Which weights? Starting weight? Finishing weight? Spot weight at the start of the climb in question? Spot weight at the end of the climb in question?

                    • Daniel Andrews

                      No kidding and neither does sky, this is an assumption everyone has to make. However, we can regress all this data to the line of best fit which will give us the a reasonable estimate of every single rider we have data points for. The thing is you want certainty and the truth is Sky’s power meter is no more a direct measure than our estimates as it’s an estimate within estimate already.

                    • lickspittle

                      And yet without so much as mentioning margins of error you are willing to insinuate that cheating is occurring. Shoddy.

                    • Daniel Andrews

                      Where did we say cheating was occurring? The best you can do is say we imply that cheating still exists, but that a probability problem that we know has improved. We would like to continue that way as too many guys are still in the final selection on climbs.

                    • Daniel Andrews

                      We deal with modeling performance, we know there is going to be errors and the likelihood is we will run into compounding error at some point but we haven’t yet. All we are asking for is more data and not just from Sky.

                  • lickspittle

                    Hardly. It’s called pseudoscience. You co-opt the language of science without bothering with the rigour. The day I see a graph with error bars on your site or Tuckers I might begin to think you’re serious.

                    • Daniel Andrews

                      Do you even understand what you are saying? Do you know the difference Science and Engineering? Modeling is part of the engineering domain as it is application of known knowledge. Right now, we don’t need to use bar graphs with error. Models have to accept variation and tolerate error to be valid. We know there is error, but we also know we have historical markers we can benchmark against as well. In systems engineering this is called fault tolerance. We know the possibility exists that we could compound errors with such a model. What makes a model valid is if it doesn’t produce results which run counter to what is possible in the real world. So far that hasn’t happened and why using our data Froome has to have a higher W/kg than Gesink up the climb and it’s why we have to reject Team Sky’s interpretation of their data and anyone with a science background unwilling to see this has lost all objectivity and No that doesn’t mean Team Sky is cheating.

                      https://en.wikipedia.org/wiki/Scientific_modelling

                      I hate to have break out the wikipedia reference, but it’s obvious you can’t follow what we have been saying and sending to a more advanced site will only confuse you more.

                      With regards to rigor, who had to find the manufacturers recommendation on power meter correction? It wasn’t Team Sky, it was Veloclinic. Do we try to verify our data with other forms of measurement? Yes, we do. Did Team Sky verify theirs? No! Do we try to use the best data available? Yes. Do we use our own data? Yes. Do we ask questions when data doesn’t fit? Yes.

                      Here is the other real world problem with Sky’s data.

                      The more Gesink and Froome weigh the less likely either of them would have climbed as fast as either of them did as excess mass becomes detrimental to performance the longer the performance takes and the lower the W/kg the slower the ride.

                      If you believe Gesink weighs between 73-75kg and Froome weighs between 67-68kg at the end of those climbs so say. Gesink @ 70kg is 5.84, @ 73kg is 5.60, and @75kg is 5.45, because that’s the range his mass would have to be with Froome 67.5 @ kg to lose 93s on that climb using Sky’s data while the regressed data matches up well with a 65 kg Froome and a 70kg Gesink. That’s not psuedo-science, that’s called an estimate with a range and at 5.78 W/kg it’s not a likely scenario that Froome climbs as fast as he did. 5.98?? That’s feasible, 6.10 that’s feasible as well and also not subject to being Pantani or Contador (Verbier) class material, but definitely a little on the high and fast side. Speed, Mass, and Work Required are the 3 inescapable things here, all 3 are known with reasonable accuracy with only being used as a range.

                • Paul Jakma

                  The thing is, the way the calculations work, the weights ultimately don’t really matter much. A heavier rider will need to use more energy to climb, but they’ll produce more power. The lighter rider produces less power, but weighs less. These effects cancel each other out to a significant degree, given the range of variation. So you don’t have to be spot on with the weight – you just need to be within a kilo or three. The resulting W/kg figure will be relatively accurate, regardless of whether the rider is lighter, equal to, or heavier than the assumed mass.

                  In short: it’s all about the W/kg – the precise weight of the rider isn’t that critical to the calculation. The way the calculations work, you just have to be in the ball-park of the actual weight and you’ll still get an accurate W/kg number.

            • Daniel Andrews

              Yates data is online. Pinot’s data is online. Gesink’s data is online. No one is making anything up. We aren’t even accusing Sky of making things up, but don’t give us Froome’s morning weight 7 days after a climb on a rest day and make sure you use a correction the manufacturer says should be used on the power meter when osymetric chainrings.

      • lickspittle

        Or maybe you think that the likes of Tinkov and Vinokourov would keep shtum for the good of the sport?

  • Superpilot

    I also just want to say, this talking head has used the finishing time gap for the stage, rather than the actual time for the climb upon which the data is based, in order to compare Ten Dam, Yates, Gesink. Gesink blew after hitting out, which obviously would reduce his overall w/kg. Just pick the data and numbers that suit you, and continue hating on cycling, mate!

    • Daniel Andrews

      If you and I race a climb on a bike and we finish in the same time we will have used the same relative power to make it up the climb. If one of us is faster than the other then the one who finished fastest will have used a higher relative power to do so. BTW, The finishing gap for the stage is the same as the finishing gap for the climb.

      • SteveC

        Can you just explain that last sentence to me please?
        If the finishing gap for the stage between rider A and rider B is 4 mins, then the gap for the climb is also 4 mins even if rider A got to the climb 4 mins before rider B?

        • Daniel Andrews

          The last sentence in this case is a reference to Gesink/Yates v Froome data argument. All of whom started the final climb at the same time, but had different finishing times the finishing gap is the same as the climbing gap, but Froome who was faster than both sits between both in terms of relative power output using Sky’s interpretation of their own data.

          • SteveC

            Ah, OK, I understand now. What you meant to say was “BTW, The finishing gap for THAT stage WAS the same as the finishing gap for the climb.”

            Also, does anybody know the difference in efficiency of high vs low cadence? Sure I heard somewhere that getting out of the saddle is not as efficient. Rather like HP at the wheel is not the same as engine HP, same model of car with the same engine HP will give different HP at the wheel.

            So would a rider out of the saddle use more energy to give the same power at the pedals?

            • Daniel Andrews

              efficiency doesn’t matter in this calculation. Efficiency though would be a big help in a larger picture of having power profiles.

              • SteveC

                Would putting more torque into the bottom bracket because of a slower cadence reduce the overall efficiency of the drive chain due to extra load and flexing?

                Also, if you are seated wind drag is a lot less than standing, so could a rider with less wind drag ride faster with a lower power input than a rider with more wind drag?

                I know they are got a “lot” slower than a sprinter but there is a difference between the power input for Cavendish and Greipel for the same speed.

                • Daniel Andrews

                  The only sprinter who uses non-round chainrings i know of is Greg Henderson. My guess is you sacrifice low end torque for a smoother power profile during the pedal stroke which costs you the initial impulse to sprint effectively. Henderson is a lead out guy mostly so a smoother power profile probably allows him to maintain speed over a slightly longer distance, but at the cost of peak power. I hope that makes sense and is correct, but a lot of this would have to be researched and verified. Again if we had complete transparency this becomes the sort of thing we could gain insight from.

  • PaulC

    I’m not usually inclined to join in these internet bun fights but reading that article I’m afraid anyone that says “It’s like a world where a car travelling at 60 is faster than a car travelling at 70.” as an analogy clearly can’t be trusted to analyze the figures. That’s complete rubbish. Its more like say a car pushing out 60BHP is going faster than one pushing out 70BHP, oh yea that obviously could be the case with the weight of the car and aerodynamics etc. Sure Froome has questions to answer just like everyone else in the sport but Mr Tucker is not the one to analyse the data with statements like that!

    • Daniel Andrews

      On a climb, power = speed and vice versa.

  • Nath

    The whole argument places far too much emphasis on power. It assumes each rider is at their maximum and that no other factor contributes. Riders are not machines and are effected by the weather, tactics, injury, illness and self doubt – all to varying degrees and at different times. That can result in some riders not performing to a hypothetical maximum this large gaps.

    Also, those hypothetical maximums are based on mathematical analysis of past performances, not a scientific study of what is actually physically possible. It is assumptions based on assumptions based on assumptions.
    I am not suggesting anyone is clean or dirty, but making any determination on power data alone is in no way valid or logical.
    We also need to realise that we (the fans/internet) are not the arbiters. If it was me, I would be sending my power data and blood work for all my races and training to the UCI or WADA for analysis and no where else. Trial by fan is just stupid and proves nothing either way.

    • Daniel Andrews

      unlike on the flats, climbs is about man against his own mass and that of the bike (which all happen to be about the same mass w/ a .2 kg total variance). They all climb with the lightest bike the UCI regulations will allow, not doing so would be stupid.

      unlike the UCI and WADA, the internet has capable people like Dr. Tucker, Vayer, Veloclinic, and myself who understand the physics involved, the need to be open and transparent, and have all researched human performance especially related to power, power duration curves, and performance modelling.

      • Rich

        “unlike on the flats, climbs is about man against his own mass and that of the bike (which all happen to be about the same mass w/ a .2 kg total variance).”

        Unfortunately, this is wrong, as playing with any bike power calculator will show you. While it’s true that the majority of the power goes towards overcoming gravity, these guys are going around ~20km/hr a lot of the time and at that speed the power needed to overcome air resistance etc. is around 60W — some 15% of the total power. So if we take a 70kg rider with a power-to-weight of 5.8W/kg versus a 60kg rider putting out 6.0W/kg, the former is generating 406W and the latter 360W. Taking off the value for air resistance the 70kg rider is putting 346W into overcoming gravity (=4.38W/kg including 9kg for bike and kit), whereas the 60kg rider is putting 300W into it (=4.35W/kg for the same assumption). The 70kg rider thus pulls away from the 60kg rider, despite a lower power-to-weight.

        Neglecting this bit of the calculation makes Tucker’s comment ridiculous.

        • Daniel Andrews

          This wrong on so many levels and there are studies which show this not to be the case, plus the CPL model accounts for wind resistance. The Dr. Ferrari model does not and both come out similar. The reason is simple we are measuring actual physical work and not efficiency. Besides your 15% argument is a bit off especially when related to climbs and cross-sectional surface area + speed is what accounts for aerodynamic drag. The rider with the larger mass is going to have a larger cross-sectional surface area. This is the beauty of normalizing power with respect to mass on uphill climbs.

          http://www.naspspa.org/AcuCustom/Sitename/Documents/DocumentItem/2415.pdf (CPL model has roots here)

          http://link.springer.com/article/10.1007/BF00236708

          http://www.researchgate.net/profile/Gregoire_Millet2/publication/11088358_Level_ground_and_uphill_cycling_efficiency_in_seated_and_standing_positions/links/0c960526a507651e86000000.pdf

          http://www.carlocapelli.it/ftp/Didattica/Specialistica/Articoli_Lavoro_Studenti/24.04/Olds_et_al_1993.pdf

          • Rich

            I will merely gently point out here that if, as you claim, the wind resistance terms nicely cancel out, then flat stages would all be about the power to weight as well. But they’re not. The obvious conclusion is that there is a simplifying assumption being used – which only makes a few percent difference, but we’re only talking about a few percent. It seems pretty obvious that the rider with the higher absolute power, for the same power to weight, will go uphill faster since they will have less trouble overcoming air resistance.

            • Daniel Andrews

              for climbs of 3km or more in length it is generally accepted as 4% in grade. For really short and really steep climbs like the Mur de Huy and Cauberg that last less 3-4 minutes and take longer than 45s it comes to down anaerobic capacity. Guys like Gilbert, Van Avaemet, Cancellera, Martin, Sagan, Kwiatkowski, and all can perform against the pure climbers if they are in the right gear ratio. That’s not saying the heavier TT specialist or classics rider shouldn’t be able to defend position on all Cat 2’s or most Cat 1’s as they shouldn’t be distanced that much on those climbs and it really depends on the steepness. The stage tomorrow is one of those days where a TT or classics specialist would be able to defend or even win possibly in a one-day or a short stage race. A classic example would be a rider like Stybar on Tomorrow’s/today’s stage.

          • Rich

            Or perhaps instead, you could tell me exactly what gradient cycling suddenly stops being about absolute power and becomes about power to weight? When does it start counting as a “climb”?

      • Nath

        Sure – going uphill limits the effects of wind resistance, but it doesn’t limit the impact of the other factors I mentioned i.e. weather, tactics, injury, illness and self doubt. All of these things effect the power being utilised. Just as a poor mental outlook will limit the max and average power output, so a very strong positive outlook can increase the possible power output. Is there any study that hypothesises maximum/average human power output that isn’t based only on estimates of past efforts? I am yet to see one. Endurance athletes are constantly redefining the physical limits of the body and much of this improvement is about the mind.
        As mentioned earlier, I am not saying anyone is or is not doping. All I am saying is that using power data as the ultimate and sole arbiter of clean performance is not a valid scientific approach. At a past national level junior points race, my son successfully made a break and gained a half lap on the bunch. He was being chased hard by his mate who was unable to make a dent on the advantage. My son is a mid pack racer at best and his mate is the second best in the country – their power numbers are vastly different, but in those few minutes my son rose above the data and his own history and did something that for him was extraordinary. I know this isn’t an example to be applied directly to the situation with Froome et al, but I think it shows that power data is not the only factor in racing.
        On a side issue – interesting comment re “capable people”. The problem with these capable people is that are not authorities within the process, rather they are mere commentators, informed as they may be. I would like to see these people or similar actually employed by UCI/WADA within a legitimised process, but even then there will be questions, just as there are when recognised experts analyse the bio passport. The problem is that the analysis of data and sometimes the data itself is subjective. As such, no matter what the experts come up with, there will be an equally informed contrary view. To some problems there are no solutions.

        • Daniel Andrews

          Those are psychological and physiological variables, except for weather which is the same for everyone. The models we use are entirely about mechanical demand and demand based models. There is data and research which back up physiological demands of the supply side to achieve these mechanical outputs?

          http://journals.lww.com/acsm-msse/Fulltext/2011/05001/Critical_Power_And_Aerodynamic_Drag_Accurately.488.aspx

          Does British Cycling using the same models change your tune?

          • Nath

            Not sure what ‘tune’ I am changing? I think you have misunderstood my position entirely. I am not arguing the science of power or the models used but rather the validity of the conclusions…I am merely pointing out that there is more to performance than power data alone. If physiological and psychological variables are not taken into account then surely the power numbers alone are not a definitive indication of ‘other-worldly’ performance and therefore the whole discussion is largely pointless. btw – weather effects people very differently, the same weather does not effect each rider equally.
            Lastly, you refer to yourself as a “capable person” but a quick search of your name along with sports science and a number of other terms brings up this article and…. crickets. Do you have some sort of expertise in this field? Genuine question.

            • Daniel Andrews

              Expertise? Hard to quantity that, but experience is yes I have that. BS in Physical Education with Emphasis in Biomechanics and Motor Control. Graduate coursework in Biomechanics, Neurobiology, and Endocrinology. I am also a collegiate XC/track and field coach and have coached for 12 years, a contributor to Elitetrack.com. As a coach, I have done a lot of modeling, know what physiological and psychological variables may do to a model. The one the thing the models won’t do is break the physical laws of nature that we know and understand even when the physiological and psychological variables are the absolute worst they have ever been or the best they have ever been. The disconnect here is we are not trying to explain what is going physiologically and psychologically by asking for data, we are only trying to ensure validation of the models. We already know the physiological variables needed to perform at certain relative power for a certain given unit of time. We also know it’s possible to have 1 time peak performances which can put an athlete at or near the mutant category of human performance, we also know how long roughly someone who is well rested can hold and maintain peak performances multiple times in a given season. These are things we understand and know already, some of them have been modeled well. I try to use psychological batteries to account for psychological variables relating to stress in my own athletes when modeling their performances. Is it perfect? No, but something I have carried from my other profession as a software engineer is including fault tolerance when modeling and staying away from models which don’t.

  • Push Bike Writer

    Good update piece Shane. Here’s my thoughts published this morning on why Chris Froome deserves more respect, and why Paul Kimmage and others are wrong to insist on the public release of rider data http://theconversation.com/chris-froome-the-tour-de-france-and-cycling-deserves-better-45053
    Craig Fry

  • Aaron J HS

    This is all just back and forth, inside out and upside down.

    On the one hand, to be fair to Tucker in his original blog he does state that he doesn’t believe that analysis of these particular numbers proves doping and he states that he doesn’t feel that even his adjusted figures are all that “mutant” or “otherworldly” to risk misquoting him I’m sure he said something similar to it was consistent with the gaps between the contenders. Also he does concede that he doesn’t KNOW the weights of Gessink, Yates and Ten Dam but that the difference in they’re uploaded data were consistent enough to demonstrate the time gaps between them to suggest that their reported weights were accurate. His point is merely that he wants transparency and this is not transparency.

    But on the other hand the day after the stage 10 he posted a blog with power to weight ratio ESTIMATES of Froome that far exceeded those that he now estimates, he states he used several forumlae to come to these conclussions. However at that time he knew at least Gessink’s and Ten Dam’s power numbers and knew the time gaps. To now say that the 6.10w/kg figure is consistent with the time gaps to Gessink,Yates and Ten Dam but just a week before to feel that a Froome was doing a 7+w/kg ESTIMATE when he knows how far ahead he finished from GESSINK and knows GESSINKS’s numbers is a little odd. So take from that what you will.

    Personally while it’s frustrating as a viewer to see the yellow jersey really under no pressure and to see the race effectively nueteralised between the main contenders. Although my initial emotional reaction is that he must be doping, my common sense approach is that there is way to much to consider to make that assumption. He only appears superhuman compared to the guys he’s leaving behind. Nibbles was dropped well before Froome attacked, Contador is clearly low on energy after the Giro, Van Garderen and Valverde have always struggled to keep up with the top contenders and that only leaves Quintana. Realistically Quintana is the one rider of the contenders we know the least about, he’s still very young (so we can’t accurately assess his race tactics and consistency), he trains at home in Columbia (so we don’t know if he’s getting the work done that he needs to) and it was the day after the rest day (we don’t have enough information to know how that would affect him). Is it a surprise that Froome is leaving behind a 23 year old that we know so little about? Not really. The fact Froome slowed down to slower than he was pre attack and kept riding away from Quintana whilst also the fact Porte was able to catch and then attack Quintana without a reaction, suggests the guy was just on an average day or in a situation that he wasn’t quite sure how to control.

    So I think that stage itself is no indication of whether Froome is doping. Where he found that bottomless box of matches in the 2011 Vuelta right when he was desperate for a new contract is another debate altogethor.

    See what I mean? Upside down, inside out and back and forth.

    • Dave

      Excellent post.

    • Tom Wells

      Seconded, great post.

  • Jake(Aus)

    I do worry that if we go to far with checking everyone’s data to the point that we know every riders top ability, it will get to the point where it’s like why bother racing? We will know on any given climb precisely where they should all finish within one another. Obviously there will always be variables like fatigue and racing strategy on every given day, especially for a sprinters stage but my point is getting right down to this level of detail in knowing someone’s absolute max ability in what is basically a flat out endurance sport, especially for GC grand tour situations, kind of take the fun out of the whole thing doesn’t it? It’s like riding with that mate you know is just better than you and who will beat you up a given climb any day of the week, after a while you just accept it and see them up the top knowing 99% of the time they’ve got ya covered. I can picture some table in future showing everyone’s max data and we can just order them then and say yep genetically rider A is the best, just give him the trophy. You’ll all talk about training etc. but I’m talking amongst the best pro teams here so all things considered let’s face it, the GC winner is the one with the best physiology.

    • Dave

      “You’ll all talk about training etc. but I’m talking amongst the best pro teams here so all things considered let’s face it, the GC winner is the one with the best physiology.”

      Rubbish. Skills and tactics still play an absolutely huge part in bike racing – remember last year when Froome and Contador lost the Tour based on their skills rather than their physiology? What about Thibaut Pinot who loses time whenever the road is not going uphill?

      There’s plenty of room to increase the importance of skills and tactics, I can think of three simple fixes:
      1. More interesting courses which reward racers rather than just one-dimensional pedalling machines. This year’s Tour has already gone that way to some extent, the first week with the seven different stages in seven days was fantastic.
      2. Ban two-way radio between riders and team cars which is used for coaching on the road. Keep radios just for Radio Tour and for riders to make one-way requests for assistance which could even be a number of buttons to send text messages (e.g. mechanical, food/drink, medical, crash) rather than vocal transmissions.
      3. Ban the in-race use of power meters by riders, put the head units under the saddle like they are for track cycling so they can only be used to analyse the data once the day’s racing is over. This will shift the emphasis from having the best physiology to having good physiology and knowing how to use it.

      • Jake(Aus)

        You lost me at ‘Rubbish’…lol sorry expert. You’re right if you plan everything out right and train right hey what’s physiology. I think most times the winner is the one who is putting out the best power to weight ratio during whichever years tour. Is that you opinion? no. Fine you are entitled to it, I wouldn’t call it ‘rubbish’, it’s just an opinion. Also your ‘fixes’ umm yeh well they aren’t in place right now so totally invalid to the discussion about whether physiology overrides tactics at the moment in sorting the GC winner. I never said there wasn’t more room to tilt in such a way that tactics had more weight.
        The one upmanship on here is getting a bit annoying you can barely voice an opinion without someone absolutely slamming you and then carrying on with their self image crafting attempt at letting the world know how superior they are.

  • Jd

    Really guys, these different people are using different petcentage deductions for the chain, AND DIFFERENT WEIGHTS for froom. Nobody has mentioned the weight loss during a race. I play squash and can lose up to 4 lbs in 40 minutes. Do the maths.

  • Cal C

    Just power to weight ratio tells only a small part of the story.

    Why not discuss the weight that is rotating- ie their legs? You know instinctively that riding with a backpack containing a few kilos won’t slow you down much at all, but attach a few hundred grams to each of your ankles and you’ll certainly notice it.
    Think about undoing a stuck nut with a 12″ spanner or undoing the same stuck nut with a 6″ spanner. Which would be easier? Obviously, the 12″ spanner requires much less power to turn the nut. So riders with longer femurs have less effort to turn their pedals when climbing. The ratio of femur to tibia can vary considerably for some it’s 1 to 1 and a few people (esp. climbers) are 1.3 to 1.

    • AllanO

      Well, no actually, to undo the stuck nut at the same rate (speed) requires exactly the same power with the 12″ spanner and the 6″. Just more force with the 6″. Basic physics.

      You might as well say it requires less “power” to climb in (say) 34/28 as it does in 39/25 at a given speed.

  • mitch_brooklyn

    This is a total waste of time.

    What about rolling resistance? Tire choice can be serious watts.

    Sitting or standing? On a wheel, at the front or all alone. All result in aero differences at these climbing speeds.

    L/R imbalance with Stages?

    Power measured at the crank, so what about drive train loss? Wax or lube? That’s watts too.

    I would totally dismiss this as chauvinistic French whining, except they turned out to be right about Lance….

  • Dave Towse

    Who knows? Ultimately, for what feels like the millionth time, the
    performance data cannot and will not provide PROOF of doping, or no
    doping. The rest is just looking for worst in everyone to suit your angle – I’m in the press or I’m a pseudo scientist who wants publicity for my own financial reasons

  • Trev

    Can you just do the math on the osymetric riders? I doubt the manufacturer could know what gains someone using them for years tuning their technique would achieve

BACK TO TOP

Pin It on Pinterest

14 NEW ARTICLES
December 5, 2016
December 3, 2016
December 2, 2016