Power Data Estimates for the climbing stages

[mr.biggles]

Van den Broek is practically solo.

Lotto having a crappy team in the mountains isn't exactly evidence of a clean peloton. It's evidence that you're watching the Tour de France.

 

[Cozy Beehive]

Anyone ready for John Wilcockson's speculation of Armstrong's power to weight ratio during his attack on stage 16?

On the Peyresourde, Sørensen was with his team leader Andy Schleck in a small group that went over the summit a minute behind Armstrong’s breakaway; so the Texan had to have been working even harder, perhaps hitting 6.5 watts/kg, equivalent to averaging 450 watts for the half-hour climb. And that was just the start of a near-six-hour stage over four mountains.

No wonder the peloton was ripped apart!

http://velonews.competitor.com/2010/07/tour-de-france/inside-the-tour-with-john-wilcockson-everyone-needed-a-rest-after-that-stage_130539

 

[131313]

Anyone ready for John Wilcockson's speculation of Armstrong's power to weight ratio during his attack on stage 16?



http://velonews.competitor.com/2010/07/tour-de-france/inside-the-tour-with-john-wilcockson-everyone-needed-a-rest-after-that-stage_130539

I didn't see Sorensen's power file yet, but I saw the screenshot of Horner's. 5.9 w/kg for 30 minutes in the front group, a minute down. Doing the math in my head, and without seeing the stage, I'd guess there's a 6.5 w/kg 5 minute section in there, with an overall of roughly 6.2 w/kg. For 30 minutes, that's certainly a lot, but doable, particularly on the first climb of the day.

It's a far cry from 6.7-6.9 for almost 50 minutes, that's for sure.

 

[131313]

Anyone want to stake a stab on Horner's power going up the Tourmalet? It would be interesting to see how close people can come before the data are online.

Based on his other efforts of the race, I'll say 340W, which is about 5.4 w/kg.

 

[Parrot23]

Yes, would be interesting as he was close to the front: chance to compare a good baseline set of data vs. estimates near leaders.

 

[131313]

Yes, would be interesting as he was close to the front: chance to compare a good baseline set of data vs. estimates near leaders.

well, already knowing Horner's power data throughout the race, it makes this question way easier to answer. Still, I'm curious how the models compare to reality. That, of course, is assuming that his SRM is calibrated.

 

[djconnel]

VAM, anyone?

Here's a profile.

 

[DarkWing]

the VAM for today?
I timed andy and contador at about 49minutes.
and they climed 1405 meters.
that's a VAM of 1720

 

[djconnel]

Okay.... then some quick assumptions:

mass = 65 km
bike = 7 kg
equipment = 2 kg
total = 74 kg
Crr = 0.3%
CdA = 0.35 m²
rho = 1.1 kg/m³
distance = 19 km
net climbing: 1404 meters
time = 49 minutes
drivetrain efficiency = 98%

6.35 W/kg powertap
6.48 W/kg SRM

Compare with L'Alpe d'Huez VAM numbers according to Science of Sport blog


And "Bjarne Riis is estimated to have produced 6.8W/kg (480W) on Hautacam when he won the Tour in 1996. Armstrong's estimated power output on Alp d'Huez was 6.6 W/kg (465W)..

No sign of special sauce.

 

[smaryka]

From the look of the flags, there was a tailwind for at least some of it.

Regardless, some people had a very good day after the rest day (Horner) and others not so good (Leipheimer).

 

[djconnel]

In my estimate, 0.80 (powertap) or 0.82 (SRM) W/kg were wind resistance.

So even if you assume each rider drafted 50% of the time and that draft was worth 25% of wind resistance, that reduces the estimate by around 0.11 W/kg, to 6.37 W/kg SRM.

Toss in a tailwind, and it the error bars start to blow out.

But we're definitely well on the clean side of 6.7 W/kg.

 

[Cozy Beehive]

the VAM for today?
I timed andy and contador at about 49minutes.
and they climed 1405 meters.
that's a VAM of 1720

53 minutes is what my 1 hour footage that I recorded says.

 

[Cozy Beehive]

VAM, anyone?

Here's a profile.

On the conservative side, with 1700 m/hr. Analysis here.

 

[djconnel]

Cozy Beehive has a slower rate of climbing, with some really nice analysis.

Was the 49 minutes from Contador/Schleck, or from the breakaway?

Anyway, power appears to have been even lower than I estimated.

 

[Le breton]

Cozy Beehive has a slower rate of climbing, with some really nice analysis.

Was the 49 minutes from Contador/Schleck, or from the breakaway?

Anyway, power appears to have been even lower than I estimated.

I am waiting for the SRM data to be published. However I am pretty certain that the time taken by Schleck-Contador to climb from LUZ ARDIDEN SAINT-SAUVEUR, SORRYcenter ( town hall usually for such places) at an altitude of 710-711m to the top of Tourmalet (2115m) was above 49min 10 s. and definitely below 50min. The distance between those two points is about 18.7km. I just noted down the times on a piece of paper at various points, like 20km, 15km, 10km etc, as well as GPS distances.
I expect the time to have been approximately 49min 30s.

Anyway, Portoleau knows very well that area and was probably on site.

Thanks nonetheless to Cozy Beehive for the nice work, maybe he can correct his time data for Schleck-Contador

Cheers

PS : it looked like the finish line was a little bit lower than the actual summit of the road.

 

[Ferminal]

In my estimate, 0.80 (powertap) or 0.82 (SRM) W/kg were wind resistance.

So even if you assume each rider drafted 50% of the time and that draft was worth 25% of wind resistance, that reduces the estimate by around 0.11 W/kg, to 6.37 W/kg SRM.

Toss in a tailwind, and it the error bars start to blow out.

But we're definitely well on the clean side of 6.7 W/kg.

Isn't that still quite high for the 18th day of a GT?

 

[Cozy Beehive]

Thanks nonetheless to Cozy Beehive for the nice work, maybe he can correct his time data for Schleck-Contador

Cheers

PS : it looked like the finish line was a little bit lower than the actual summit of the road.

I timed all 18 sections of the climb, every kilometer. Now if it really is 49 minutes, that's a huge difference (3 minutes!). I do have some doubts but you're welcome to get official figures for me so I can change it.

For a rough estimate, look here. The riders passed the 18K to go mark at 3:40 CEST. When Contador and Schleck got done at the summit, the time was 4:34 CEST. Count on your fingers. 54 minutes. My time in the analysis 53 min 25 sec. What am I doing wrong now?

 

[The Science of Sport]

Ross, I'm not going to hijack your topic. Go ahead. Speak your mind. I am interested in finding out how altitude or wind had a role to play in these numbers. Would these men show the same figures at or close to sea level?

Ah, but you should!

Your analysis of yesterday's Tourmalet stage was excellent. Here's the link:

http://cozybeehive.blogspot.com/2010/07/tour-de-france-stage-17-col-du.html

Sure, they're estimates, but the approach is great, and I think your figure of 6W/kg is about right, it corresponds to what I would have calculated using a much more simple method. As you say in the post, Horner's data will make for interesting 'validation'.

Nice job!

Ross

Edit: Sorry, I see the link had already been provided by page 7 - I responded, then checked. My bad.

The actual climb - a mistake on the road?

And just another thing, for everyone calculating the VAM yesterday based on a climb of 18.6km at 7.5%.

On the TV footage, the 18.6km to go banner has to have been in the wrong place. It wasn't 18.6km. You know this because the section from 18.6km to go to 15km to go (3.6km) was covered in about 6 minutes, a speed of 36km/hour!

The climb from that point on was done at about 21 km/hour. So yes, they were fast entering the climb, but not almost twice as fast as for the rest of the climb...

Also, from the 20km banner to the 18.6km banner, it took 4:42 (a speed of 17.8km/hour). And that was on a relatively flat section.

So clearly, that 18.6km to go banner was not correct. I would estimate that it was about 1 km late. Which is why I agree more with Ron's calculation that the 18.4km took about 53 minutes and not the 49 min that I timed from that banner.

That 49:10 that some of you have timed (I got it at 49:08) is from the banner, right? Well, I think the banner was actually about 17.6-17.8km from the finish, and the gradient should thus be 7.6%.


if you do this, then you get can estimate a power output for the whole climb of 5.94 W/kg, which is mighty close to what Ron worked out for the final part after the attack. I think this is a reasonable figure, but Horner's data later will shed some light.

So, bottom line - about 6W/kg, when clearly race strategy dictated that the very best riders give it everything they can today. The (correct) argument about power outputs being lower because of race situation can now be used to say that today was a maximal effort, and still, they're only at 6W/kg. Compare the 6.2W/kg to 6.5W/kg of the 90s and early 2000s. A good sign for cycling...

Oh, and finally, also look at the time gaps between Schleck and Contador and the chasers. It grew quickly to a minute, then edged out to 1:30, and then stayed pretty much there. A sign that the attack, which was fierce, was above the level that even the very best climbers in the Tour could sustain.

 

[Le breton]

I timed all 18 sections of the climb, every kilometer. Now if it really is 49 minutes, that's a huge difference (3 minutes!). I do have some doubts but you're welcome to get official figures for me so I can change it.

For a rough estimate, look here. The riders passed the 18K to go mark at 3:40 CEST. When Contador and Schleck got done at the summit, the time was 4:34 CEST. Count on your fingers. 54 minutes. My time in the analysis 53 min 25 sec. What am I doing wrong now?

I have climbed Tourmalet only twice, 20 years ago, and in the fog or under clouds, so that I don't know it as well as, say Alpe d'Huez. Therefore I don't know where exactly is located the 710m altitude point.

The watch I used is 2min46s fast, to compare to real time you need to subtract 2:46 from indicated times.

Here are my notes
16:39:39 leader at -20km (Contador-Schelk at 4:04)
16:42:17 leaders at -18.7kmGPS(C-S at 4:00)
16:43:40 last of C-S pack at -20km
16:46:17 leaders at -16.7kGPS (C-S at 3:07)
16:48:00 leaders at-16km GPS (C-S at 3:00)
16:50:11 leaders at -15.0km GPS (C-S at 2:43)
16:52:54 C-S at -15km
16:56:34 leaders at -13 km GPS (C-S at 2:26)
17:02:50 leaders at -11km GPS (C-S at 1:44)
17:05:50 leaders at -10km (C-S at 1:20)
17:06:50 C-S at -10km
17:09:37 leader at -9km GPS (C-S at 7s, in fact at that instant they were just 5m behaind)
17:23:50 C-S at -4km
17:35:58 C-S at finish.

I had other info on a chronometer but accidentally erased it.

SO : between leaders passing -18.7 km mark in Luz-st-sauveur and C-S at the finish line, 53:41 elapsed. But C-S were about 4:00 in Luz, therefore their time up Tourmalet was close to 49:41

Hope this helps

 

[Le breton]

Ah, but you should!

Your analysis of yesterday's Tourmalet stage was excellent. Here's the link:

http://cozybeehive.blogspot.com/2010/07/tour-de-france-stage-17-col-du.html

Sure, they're estimates, but the approach is great, and I think your figure of 6W/kg is about right, it corresponds to what I would have calculated using a much more simple method. As you say in the post, Horner's data will make for interesting 'validation'.

Nice job!

Ross

Edit: Sorry, I see the link had already been provided by page 7 - I responded, then checked. My bad.

The actual climb - a mistake on the road?

And just another thing, for everyone calculating the VAM yesterday based on a climb of 18.6km at 7.5%.

On the TV footage, the 18.6km to go banner has to have been in the wrong place. It wasn't 18.6km. You know this because the section from 18.6km to go to 15km to go (3.6km) was covered in about 6 minutes, a speed of 36km/hour!

The climb from that point on was done at about 21 km/hour. So yes, they were fast entering the climb, but not almost twice as fast as for the rest of the climb...

Also, from the 20km banner to the 18.6km banner, it took 4:42 (a speed of 17.8km/hour). And that was on a relatively flat section.

So clearly, that 18.6km to go banner was not correct. I would estimate that it was about 1 km late. Which is why I agree more with Ron's calculation that the 18.4km took about 53 minutes and not the 49 min that I timed from that banner.

That 49:10 that some of you have timed (I got it at 49:08) is from the banner, right? Well, I think the banner was actually about 17.6-17.8km from the finish, and the gradient should thus be 7.6%.


if you do this, then you get can estimate a power output for the whole climb of 5.94 W/kg, which is mighty close to what Ron worked out for the final part after the attack. I think this is a reasonable figure, but Horner's data later will shed some light.

So, bottom line - about 6W/kg, when clearly race strategy dictated that the very best riders give it everything they can today. The (correct) argument about power outputs being lower because of race situation can now be used to say that today was a maximal effort, and still, they're only at 6W/kg. Compare the 6.2W/kg to 6.5W/kg of the 90s and early 2000s. A good sign for cycling...

Oh, and finally, also look at the time gaps between Schleck and Contador and the chasers. It grew quickly to a minute, then edged out to 1:30, and then stayed pretty much there. A sign that the attack, which was fierce, was above the level that even the very best climbers in the Tour could sustain.

According to my notes the leaders did the 2 km between GPS -18.7 km and -16.7 km in exactly 4 minutes, that sounds quite reasonable for roughly 4.5% incline : 30 km/h.
http://www.climbbybike.com/fr/profil.asp?Climbprofile=Col-du-Tourmalet&MountainID=26

I should add that the GPS km data seemed to be quite reasonable yesterday. For example, at - 11km you could see that they were in Barèges. At -18.7 km that they were in Luz, at -20km that they were still outside of Luz.

For the GPS time difference data, there is always a time delay so that they can never be very accurate, at least not at better than 10s level.

 

[DarkWing]

yes my time of about 49minutes was from the banner.
I did however, notice that when the breakaway was at the 18.6km banner the graphic on the screen said about 17.3km to go.

 

[Le breton]

yes my time of about 49minutes was from the banner.
I did however, notice that when the breakaway was at the 18.6km banner the graphic on the screen said about 17.3km to go.

i.e, that there was a distance of 1.3 km between the leaders and the Schelk-Contador group I guess.

 

[DarkWing]

i.e, that there was a distance of 1.3 km between the leaders and the Schelk-Contador group I guess.

No, when the BREAKAWAY, including kolobnev, was at 18,6 the graphic said 17,3 - the graphic is for the riders furthest ahead.

 

[Le breton]

According to my notes the leaders did the 2 km between GPS -18.7 km and -16.7 km in exactly 4 minutes, that sounds quite reasonable for roughly 4.5% incline : 30 km/h.
http://www.climbbybike.com/fr/profil.asp?Climbprofile=Col-du-Tourmalet&MountainID=26

I should add that the GPS km data seemed to be quite reasonable yesterday. For example, at - 11km you could see that they were in Barèges. At -18.7 km that they were in Luz, at -20km that they were still outside of Luz.

For the GPS time difference data, there is always a time delay so that they can never be very accurate, at least not at better than 10s level.

To verify the above iI ran analyticycling for different slopes until I got reasonable power values for the leading group.
Result
Forces On Rider


Frontal Area 0.60 m2
Coefficient Wind Drag 0.50 dimensionless
Air Density 1.130 kg/m3
Weight 78.0 kg
Coefficient of Rolling 0.004 dimensionless
Grade 0.040 decimal
Wind Resistance 11.7 kg m/s2
Rolling Resistance 2.8 kg m/s2
Slope Force 30.6 kg m/s2
Cadence 100. rev/min
Crank Length 170. mm
Pedal Speed 1.78 m/s
Average Pedal Force 209.9 kg m/s2
Effective Pedaling Range 70. degree
Effective Pedal Force 539.8 kg m/s2
Speed 8.30 m/s
Power 373.7 watts
--------
Hence 30 km/h in drop position (CdA =0.3 m^2)
possible on 4% incline with less than 5.5 W/kg.
4.5 % incline gives too much power required.

Now, what power does Cancellara need to exert to regain about 50 s over those 2km as actually happen?
You will know in a minute

 

[Le breton]

To verify the above iI ran analyticycling for different slopes until I got reasonable power values for the leading group.
Result
Forces On Rider


Frontal Area 0.60 m2
Coefficient Wind Drag 0.50 dimensionless
Air Density 1.130 kg/m3
Weight 78.0 kg
Coefficient of Rolling 0.004 dimensionless
Grade 0.040 decimal
Wind Resistance 11.7 kg m/s2
Rolling Resistance 2.8 kg m/s2
Slope Force 30.6 kg m/s2
Cadence 100. rev/min
Crank Length 170. mm
Pedal Speed 1.78 m/s
Average Pedal Force 209.9 kg m/s2
Effective Pedaling Range 70. degree
Effective Pedal Force 539.8 kg m/s2
Speed 8.30 m/s
Power 373.7 watts
--------
Hence 30 km/h in drop position (CdA =0.3 m^2)
possible on 4% incline with less than 5.5 W/kg.
4.5 % incline gives too much power required.

Now, what power does Cancellara need to exert to regain about 50 s over those 2km as actually happen?
You will know in a minute

And the answer is 8 W/kg, which seems a bit high. But since the time differences are not all that precise, it could be that the C-S group only regained 40s over those 2 km at 4%, giving a more reasonable 7.3 W/kg for the front man.

CONCLUSION the slope was not quite 4.5% between indicated points, so that the actual climbing time between altitude 711 m and the top was a bit less than 49:41 for Schleck-Contador