So Suddenly the Tour is clean. Where did this idea come from

[hrotha]

His performance has definitely improved compared to last year - look at his season as a whole. However, I'd look at Sassi/Mapei for the reason rather than at Mantova. The implications might be similar, though.

 

[Mrs John Murphy]

His performance has definitely improved compared to last year - look at his season as a whole. However, I'd look at Sassi/Mapei for the reason rather than at Mantova. The implications might be similar, though.

It is something of a paradox that Cunego who makes a big thing of not being on the dope (compared to his best years) starts performing well in GT and TDS - which is clearly a sign of a cleaner race as clean riders can only rise to the top when the race is cleaner. Only for the image of Cunego as a clean rider to be undermined by him being implicated in a major doping scandal.

 

[Frosty]

Cunego has done different races this year to last. Instead of doing the classics then the Giro then the tour he did the classics and then started building towards the tour.

Last year he finished 11th in the Giro but four of those guys ahead of him were in the l'aquila break. In the tour he finished well down but lost 17 minutes after a crash on the Arenberg stage and tried stage hunting after that.

He finished second in the tour of switzerland this year but most of the big names went to the Dauphine. He actually did worse in the classics. Overall, its a higher level of performance but this could just be down to a better targetted campaign.

 

[hrotha]

It is something of a paradox that Cunego who makes a big thing of not being on the dope (compared to his best years) starts performing well in GT and TDS - which is clearly a sign of a cleaner race. Clean riders can only rise to the top when the race is clean. Only for the image of Cunego as a clean rider to be undermined by him being implicated in a major doping scandal.

Yes, it's a paradox. Personally I hope he's being questioned just because he was at Lampre and that he'll be cleared, because no matter how much I post in the Clinic, I still need to believe. A little.

 

[Mrs John Murphy]

Yes, it's a paradox. Personally I hope he's being questioned just because he was at Lampre and that he'll be cleared, because no matter how much I post in the Clinic, I still need to believe. A little.

Indeed. Cunego like Voeckler is one of the people you want to believe in and it would be a real hammer blow if he had been charging all this time.

 

[Alex Simmons/RST]

If someone could save me the time and supply a WKO+ or SRM file for Flecha's TT, along with a gpx file of the exact TT course (or a map my ride reference), and body mass for Flecha, Martin & Evans, I can run some models when I get a chance.

damn editor disappeared on me after nearly finishing long post on this.

Summary instead.

I was kindly sent Flecha's ITT power file, and I found an online map of course to obtain elevation data against which to run a virtual elevation model and segment the course (approx 100 segments).

After extensive modeling, here are my conclusions:

Flecha:
Average Power: 397W
Body Mass: 73.9kg
Power to mass ratio: 5.38W/kg
Drivetrain efficiency: 97.5% (assumed)
Bike + kit mass: 8kg (assumed)
Crr: 0.005 (assumed)
CdA: 0.263m^2
Power to aero drag ratio: 1509W/m^2

Note: a W/m^2 of 1500 is not that spectacular.

Evans:
4:13 faster than Flecha
Body mass: 7.0kg less than Flecha
Assume same weather, drivetrain efficiency, Crr, mass of bike + kit.

To ride the time Evans rode, then the following pairs of power to mass and power to aero drag ratios would apply (with average power and CdA in brackets), i.e. more power means less aerodynamic:

W/kg - W/m^2 (Power & CdA)
5.6 - 2144 (380W, 0.185m^2)
5.7 - 2070 (387W, 0.192m^2)
5.8 - 1998 (394W, 0.199m^2)
5.9 - 1930 (401W, 0.205m^2)
6.0 - 1868 (407W, 0.212m^2)
6.1 - 1811 (414W, 0.219m^2)
6.2 - 1755 (421W, 0.226m^2)

I would suggest that a CdA << 0.2 is probably unlikely for Evans (making 6.0 - 6.2W/kg entirely plausible) but I can't know for certain.

A CdA less than 0.2m^2 is very, very slick. Best I've measured is 0.185m^2 for a 60kg male on track pursuit bike (BT with double disks) who set a masters hour record and ~ 0.180m^2 for a female masters pursuit world best time holder.


Course gpx data here:
http://www.trimbleoutdoors.com/ViewTrip/1144090

Weather data here:
http://www.wunderground.com/cgi-bin/findweather/getForecast?query=grenoble,+france

 

[Franklin]

damn editor disappeared on me after nearly finishing long post on this.

Now this is more like it. My thanks for the great effort. It gives an extensive range of wattages. I assume you used weights from the official TdF medical check? How did you find out the CdA of Flecha?

I'm amused you come with a range of CdA's and extremes you encountered as this only shows that it differs between individuals. This is just one of the reasons why I dismiss mr. Coggan's insinuations.

1. It does not answer the "was the weather exactly the same" as the wind speed varied from 8 to 22 kmh and there was some rain reported.
2. The efficiency questions of Flecha's riding
3. I'd say the 8 kg for a full TT kit is pretty optimistic. Added questions: did they both have a similar wheel setup? Which brands did they use, as in TT's often rebadged speciality gear is used. I'm wondering about the aerodynamics/weight between a Pinarello versus a BMC (Walser?) bike.

A few percent (points) make a difference if added with the other questions. Considering over 6.0 watt per kg is seen as suspicious.

Had Mr. Coggan said: "The wattage per KG could be anywhere between Lemond's and Lance, so I would use caution seeing any changes" it would have been a lot less controversial.

Saying in the clinic in this exact topic ''not much seems to have changed since LA's day'' is something different. The context is quite clear. Considering mr. Coggan's stance on Lance I'm not surprised he leans towards the "not much changed" side.

 

[Alex Simmons/RST]

Now this is more like it. My thanks for the great effort. It gives an extensive range of wattages.

Yes, although I think those with CdA>0.2m^2 are the most likely.

I assume you used weights from the official TdF medical check?

No, I used Flecha's mass as recorded in his power meter file.

Evans I used a web search for that, happy to have his verified in another matter - I have assumed 7.0kg less than Flecha.

What matters is not so much the absolute power/mass/aero drag values, but the differences required to ride 4:13 faster.

How did you find out the CdA of Flecha?

That would take quite some time to explain. I used the method of Virtual Elevation modeling using Flecha's power meter file to that equate against an actual elevation profile.

Certain pairs of Crr & CdA will generate an elevation profile based on power & speed data that closely matches the actual elevation profile:

I won't go into detail here as to how it works. Where the virtual and actual elevations deviate means that there variations in things like wind, or localised changes in CdA (such as sitting up or standing when climbing) etc, but it's mostly wind.

In that sense the VE model takes into account localised variations. Sometimes there are small discontinuities due to braking.

I then use the same Virtual Elevation model to determine what it would take to ride that course under those same conditions. This is done by segmenting the course into small chunks of variable distance and gradient, and applying the rider's actual power data for those same segments to compare theoretical ride time to actual.

I then make any minor adjustments to assumptions to tweak it accordingly to match actual ride time. From then on it's a matter of comparing what it takes to go faster or slower by the set amount (in this case 4:13) by adjusting any of the variables I so please (power, mass, Crr ,CdA, air density, wind, drivetrain efficiency). By putting in Cadel's mass, we can then hold that constant, and hence change power to show different W/kg and hence calculate the CdA or W/m^2 required to ride that 4:13 faster.

For a fixed mass and ride time, if W/kg goes up, it means the rider is less aerodynamic. I am quantifying by how much for that specific course.

I'm amused you come with a range of CdA's and extremes you encountered as this only shows that it differs between individuals. This is just one of the reasons why I dismiss mr. Coggan's insinuations.

Well IMO, the CdAs required to ride 4:13 faster (or in case of Martin 4:20) suggest to me that Andy's initial swag wasn't that far out. If I had to make my best estimate, I'd say more likely 5.9-6.0W/kg.

6.2W/kg still provides a totally plausible CdA value for Evans (and his set up) compared to Flecha.

1. It does not answer the "was the weather exactly the same" as the wind speed varied from 8 to 22 kmh and there was some rain reported.
2. The efficiency questions of Flecha's riding
3. I'd say the 8 kg for a full TT kit is pretty optimistic.

1. No, although looking at the weather history on that page, it didn't appear to change all that much and the differences are reduced a bit because of the loop nature of the course. Point to point courses are more problematic in that respect.

2. Well that's an interesting question. It just so happens that the modeling I use was actual designed to measure "pacing efficiency", but it has other uses like what I'm doing here. So I ran Flecha's file through the pacing model.

I came up with a pacing score of 0.989 (I have a discussion paper I wrote which explains all this), but basically that ranks Flecha's ride as excellent pacing. World best pacing scores are 0.992-0.995.
1.0 is theoretical perfect pacing (doesn't happen).

It means that if Flecha had paced at world best standard, he could have saved about another 15 seconds for the same physiological effort. This is a measure of dosing out his effort slightly differently on that course under those exact same conditions. It isn't a measure of technical effectiveness.

So better pacing I'd suggest could only account for at most 15 of the 253 seconds difference between Flecha and Evans.

3. Perhaps, but I have applied the same assumption to both riders.
On that course, if Flecha's equipment was say 1kg heavier than Evans', that would account for about a 9 second difference in the calculations, or about 2 watts.

Which brands did they use, as in TT's often rebadged speciality gear is used. I'm wondering about the aerodynamics/weight between a Pinarello versus a BMC (Walser?) bike.

I discussed the weight differences above. This aero difference isn't all that relevant, since this is accounted for by the calculation of the W/m^2 required to go faster. That's the point, to ride faster, Evans (and Martin) had to be more aerodynamic.

If Evans had Flecha's aerodynamics (which he doesn't), then he'd require (on this course) around 6.9W/kg to ride as fast as he did. Evans is lighter and smaller, and so naturally has a lower CdA. The rest of the improvement is naturally being more aero shaped, better bike position and possiblly equipment.

A few percent (points) make a difference if added with the other questions.

Of course, this is an academic exercise, however I am making relative comparisons, rather than being concerned with the absolute numbers.

Considering over 6.0 watt per kg is seen as suspicious.

Is it?

Had Mr. Coggan said: "The wattage per KG could be anywhere between Lemond's and Lance, so I would use caution seeing any changes" it would have been a lot less controversial.

Saying in the clinic in this exact topic ''not much seems to have changed since LA's day'' is something different. The context is quite clear. Considering mr. Coggan's stance on Lance I'm not surprised he leans towards the "not much changed" side.

This is what I saw. Dr Coggan said nothing about Lemond or Lance or their day. I think some are reading things that just ain't there.

Flecha's TT = ~400 W/~74 kg = ~5.4 W/kg.

All else being equal, he would have had to sustain ~460 W (~6.2 W/kg) to challenge for the win in the TT.

Hmmm...thinking about it a bit, that certainly suggests that things haven't changed all that much.

 

[ianfra]

Are you guys autistic? What is the point of this amateur analysis apart from satisfying your own egos or feeding your autism?

 

[Alex Simmons/RST]

Are you guys autistic? What is the point of this amateur analysis apart from satisfying your own egos or feeding your autism?

It presents an opportunity to learn something, gain a deeper insight into elements of cycling performance. It's a blend of physics and physiology. What's wrong with that?

I have used the models to help assess whether riders have a pacing problem, the nature of that problem if they have one, where on a given course they are making pacing mistakes and how much time they could save.

For some it tells me they have no pacing problem and we can focus attention on other elements of performance improvement.

Sure, it's a little esoteric. So? I developed it during a time I was recovering from injury and had a bit of time on my hands.

 

[Tyler'sTwin]

Why is it so hard to believe Evans may have been above 6 W/kg? Isn't it normal for power outputs to be higher in a TT than on a HC climb at the end of a long multi HC stage (where he was doing like 5.8-5.9 W/kg, right?)?

 

[acoggan]

I was kindly sent Flecha's ITT power file, and I found an online map of course to obtain elevation data against which to run a virtual elevation model and segment the course (approx 100 segments).

After extensive modeling, here are my conclusions:

Flecha:
Average Power: 397W
Body Mass: 73.9kg
Power to mass ratio: 5.38W/kg
Drivetrain efficiency: 97.5% (assumed)
Bike + kit mass: 8kg (assumed)
Crr: 0.005 (assumed)
CdA: 0.263m^2

...which means that my initial estimate of 460 W was very close (amazing how well the R.O.T. works, eh?).

BTW, for point of reference Kraig Willett has estimated Armstrong's CdA on a TT bike to be 0.266 m^2 and his power for an hour-long effort to be 466 W, meaning that his power:CdA in TTs was just over 1700 W/m^2:

http://www.biketechreview.com/performance/supply/493-la-cant-do-it

(Note that Kraig's estimate of 466 W for ~1 h in the TT position is entirely consistent with Armstrong's self-reported power of 495 W for 30:47 min up the Col d'Madonne.)

 

[acoggan]

Considering mr. Coggan's stance on Lance

What stance is that?

 

[acoggan]

What is the point of this amateur analysis

"Amateur"? Alex is a professional cycling coach whose reputation has been built in part based on detailed analyses such as that he described. Indeed, the quality of this work is such that I have encouraged him to submit it to a scientific journal for publication, as it is far more sophisticated and insightful than most of the pacing strategy papers that have been published to date. Alas, I have been unable to convince him to do so, as he seems to have realized that he doesn't need to submit his work to a peer-reviewed publication to be subjected to anonymous sniping from clueless reviewers.

 

[acoggan]

Evans:
4:13 faster than Flecha
Body mass: 7.0kg less than Flecha
Assume same weather, drivetrain efficiency, Crr, mass of bike + kit.

To ride the time Evans rode, then the following pairs of power to mass and power to aero drag ratios would apply (with average power and CdA in brackets), i.e. more power means less aerodynamic:

W/kg - W/m^2 (Power & CdA)
5.6 - 2144 (380W, 0.185m^2)
5.7 - 2070 (387W, 0.192m^2)
5.8 - 1998 (394W, 0.199m^2)
5.9 - 1930 (401W, 0.205m^2)
6.0 - 1868 (407W, 0.212m^2)
6.1 - 1811 (414W, 0.219m^2)
6.2 - 1755 (421W, 0.226m^2)

I would suggest that a CdA << 0.2 is probably unlikely for Evans (making 6.0 - 6.2W/kg entirely plausible) but I can't know for certain.

Given the apparent breadth of his shoulders and his down-angled arms, even ~0.22 m^2 seems a bit optimistic to me.

EDIT: Obviously, the reason that Evans' estimated power:CdA goes down as his power:mass goes up is because the course wasn't flat. You might be able to narrow things down a bit more if you knew how long it took him to reach certain points on the course.

 

[gobuck]

Sorry to come late to the party. Was at the 1st time check of the TT which was just after a 110 deg corner, tunnel and short uphll. My observations: The weather was sunny and hot the entire time I was standing there. Cadel was faster thru that section than anyone. He was very close to going too fast thru the corner(crashing) but he made it look effortless. I saw the last 50 riders go thru there and most braked hard before the turn. He must have been very confident or reckless. The end result was positive but it could have gone the other way. Nice work Alex.

 

[acoggan]

You might be able to narrow things down a bit more if you knew how long it took him to reach certain points on the course.

Martin

Time check 1 @ 15 km: 20:12
Time check 2 @ 27.5 km: 40:26
Finish @ 42.5 km: 55:33

Evans

Time check 1 @ 15 km: 20:33
Time check 2 @ 27.5 km: 40:33
Finish @ 42.5 km: 55:40

Contador

Time check 1 @ 15 km: 20:33
Time check 2 @ 27.5 km: 41:08
Finish @ 42.5 km: 56:39

 

[gobuck]

Andy, sorry that I wasn't more clear. Cadel was the fastest thru the corner, not the 1st 15 kms.
The point that I was trying to make was that he was riding it balls out. I didn't see anyone take the same risks.

 

[acoggan]

Andy, sorry that I wasn't more clear. Cadel was the fastest thru the corner, not the 1st 15 kms.

I was just sharing time-splits in case Alex (or anybody else) wished to take a stab at refining the modeling. IOW, I wasn't trying to contradict you.

 

[V3R1T4S]

Junk Input = Junk Output

We all appreciate the analysis but I still do not understand why Flecha's power file is being used as the model; for the several reasons I brought up in my first post it is more likely than not that Flecha's power is over-reported - at the very least more likely to be off from the "truth" than Chris Anker Sørensen's.

I would be much more interested in seeing the analysis with Sørensen's file as the input. In the past his data has correlated well with other reported data.

 

[GreasyMonkey]

We all appreciate the analysis but I still do not understand why Flecha's power file is being used as the model; for the several reasons I brought up in my first post it is more likely than not that Flecha's power is over-reported - at the very least more likely to be off from the "truth" than Chris Anker Sørensen's.

I would be much more interested in seeing the analysis with Sørensen's file as the input. In the past his data has correlated well with other reported data.

So why should Chris Anker Sørensen's SRM file be so more accurate than Flecha's? Perhaps the Sky techs cannot calibrate the device, or Brailsford et al. are manipulating the raw data to make Flecha look to be a better rider than is actual?

Please advise why you see this data as so unreliable....

 

[V3R1T4S]

Please advise why you see this data as so unreliable....

I would advise you to read my first post as I asked...

There are a couple of qualms. No, it is not guaranteed his data is more or less accurate, but objectively looking at the points raised, statistically speaking, more likely than not his data would be farther off than CAS's to the "truth".

 

[goggalor]

It presents an opportunity to learn something, gain a deeper insight into elements of cycling performance. It's a blend of physics and physiology. What's wrong with that?

I have used the models to help assess whether riders have a pacing problem, the nature of that problem if they have one, where on a given course they are making pacing mistakes and how much time they could save.

For some it tells me they have no pacing problem and we can focus attention on other elements of performance improvement.

Sure, it's a little esoteric. So? I developed it during a time I was recovering from injury and had a bit of time on my hands.

I'm very impressed anyway.

I wonder how big a factor descending ability (and climbing, for that matter) would be on this quite tricky course. Gilbert was being interviewed on Eurosport during his warmup and according to Carlton Kirby who was translating Evans was "enormous" on the descents when time trialing, putting 3-4 seconds into Gilbert per kilometer.

 

[acoggan]

I still do not understand why Flecha's power file is being used as the model

I don't recall who first brought up Flecha's performance, but it initially caught my eye because he's closer in size to Tony Martin than, e.g., Sørensen. I agree, though, it would be interesting (well, at least as interesting as any of these back-of-the-envelope analyses/approximations ever gets) to analyze his file as well.

EDIT: Specifically, it was Escarabojo's semi-rhetorical suggestion to "divide all those numbers by Flecha's [. . .] weight and see what you get" that led me to do exactly that and then post this:

http://forum.cyclingnews.com/showpost.php?p=630852&postcount=189

Franklin then started in with the ad hominen attacks, leading us to where we are now.

 

[poupou]

To add some information here, on french TV we have seen a reportage on Evans training for that ITT a month or 2 before TDF. He was looking at each corner to keep the maximum of speed. He had repeated and tried many times.
I am not surprise that his work became successfull according what saw gobuck that day.