Power Data Estimates for the climbing stages

[Rechtschreibfehler]

If I am not completly mistaken Cobo wasn't in a breakaway on that stage, was he?
So either he (and Piepoli) did 6.1 or 6.5 as he came in alongsinde Piepoli.

 

[zlev11]

nah they both came out of the GC group

 

[Le breton]

Mean (average) watts calculations are unaffected by accelerations and decelerations. If Rider A produces 400 watts for 20 minutes and 200 watts for 20 minutes, their average wattage over the 40 minutes is 300 watts, the same as rider B who maintains a constant 300 watts for 40 minutes.

However, rider A has clearly worked a lot harder than rider B.

Coggan has a statistic called normalized power which attempts to take this into account. A normalized power figure is designed to indicate an equivalent psychological "cost" to maintaining a constant wattage.

Regarding wind, the very steepness of Angliru makes it less of a factor, particularly if calculations exclude the last 500 metres past the KOM point. Furthermore, IIRC there was fog on the mountain, indicative of very light winds.

Yes, VAM is likely to be highest on very steep slopes like the Angliru, as very little energy is wasted pushing air out of the way, or to road friction. Drafting has a lesser effect on the calculations than on less steep mountains, for the same reason.

What would be very helpful is if there is some public power data from one of the riders from the Vuelta, for whom an estimate can be calculated and the VAM-based calculations checked against.

+1 for your post, except on one point, this one :
A normalized power figure is designed to indicate an equivalent psychological "cost" to maintaining a constant wattage

I do not believe it is meant to be the equivalent psychological cost but the equivalent energy expenditure the rider could have sustained if he had maintained a constant power output instead of a variable one.
Also, to give credit where credit is due, it seems to me that Andy adapted Roger "1st 4-min mile" Bannister' TRIMP method based on heart rate to the avent of powermeters and calculations based on watts.

In general on a climb a racer will not come to a stop then sprint wildly, particularly if it's steep. Variations of a few watts, even say 10%, around a mean value will make little difference in the normalized power (as compared to the average power) so that in most cases the point initially raised by Escarabajo will make little difference. ( Counterexamples : Contador and Rasmussen in 2007 in the Pyrénées, Jeannie Longo and Van Morsel in AdH in 1993).
For the weather conditions I believe you saw correctly, slightly foggy atmosphere caractistic of still air, no wind. Wind blows those light clouds away. After the summit there was wind.

 

[rgmerk]

+1 for your post, except on one point, this one :
A normalized power figure is designed to indicate an equivalent psychological "cost" to maintaining a constant wattage

I do not believe it is meant to be the equivalent psychological cost but the equivalent energy expenditure the rider could have sustained if he had maintained a constant power output instead of a variable one.
Also, to give credit where credit is due, it seems to me that Andy adapted Roger "1st 4-min mile" Bannister' TRIMP method based on heart rate to the avent of powermeters and calculations based on watts.

You're right. "Psychological" was a mistake on my part, I meant to say "physiological". Thanks for the additional info on the TRIMP method and its history - that's fascinating!

 

[Escarabajo]

Mean (average) watts calculations are unaffected by accelerations and decelerations. If Rider A produces 400 watts for 20 minutes and 200 watts for 20 minutes, their average wattage over the 40 minutes is 300 watts, the same as rider B who maintains a constant 300 watts for 40 minutes.

However, rider A has clearly worked a lot harder than rider B.

Coggan has a statistic called normalized power which attempts to take this into account. A normalized power figure is designed to indicate an equivalent psychological "cost" to maintaining a constant wattage.

Regarding wind, the very steepness of Angliru makes it less of a factor, particularly if calculations exclude the last 500 metres past the KOM point. Furthermore, IIRC there was fog on the mountain, indicative of very light winds.

Yes, VAM is likely to be highest on very steep slopes like the Angliru, as very little energy is wasted pushing air out of the way, or to road friction. Drafting has a lesser effect on the calculations than on less steep mountains, for the same reason.

What would be very helpful is if there is some public power data from one of the riders from the Vuelta, for whom an estimate can be calculated and the VAM-based calculations checked against.

I agree with what you said, but that was not my point though.

I was referring to the most efficient way of working up the climb. Might save some energy and at the end you could have gone faster and still have the same average power.

The equation we use is intended for steady state conditions, here:

http://www.oocities.org/mdetting/sports/cycling.html

For climbs like the Angliru it becomes a non linear equation and an unsteady equation because of the nature of the climb. The Kinetic energy factor can be represented in more heat dissipated into the atmosphere.

As for the average power with accelerations and decelerations the same as the steady pace power I agree with you. Because you already know the power to begin with. But I am sure that two subjects having everything else equal, one going at constant pace and the other one going at unsteady pace, very irregular, and having the same time at the end of the climb, I can assure you the one of them employed more power to get to the top. Even if it is as small as 5 watts. Can be bigger depending on how inefficient you go on the climb. This is true for few exceptions. For the majority of the times different to Angliru might become negligible. It is usually neglected in the calculations for that reason.

Hey I am just pointing out something that is the energy equation. It is probably a moot point for the majority of the climbs, but for the Angliru could be a factor. I know in some papers I have seen them use this kinetic energy term just because they have a powerful computer to solve with numerical methods. So it exists. Small but exists.

Read the analogy that I already stated with the cars with the fuel consumption. One of them will consume more fuel even though they arrived at the same time, but one of them stopped several times along the way while the other one kept a steady pace. Small difference but still can be accounted for.

 

[halamala]

Vaughters just tweeted (at me ) that he thinks the numbers for Cobo + Froome here might be a tad high and they're closer to 6.0 and 5.7. And he'll try and work it out on Tuesday.

https://twitter.com/#!/Vaughters/status/110526529179623425

Do you have any new information?

 

[Le breton]

Vuelta a Espana 2011, Stage 15, Final climb, Angliru

From 12.0 Km remaining to mountain points banner 0.5 Km remaining = Distance 11.5 Km

Juan Jose Cobo

Elevation / Höhenmeter [m] : 1194 m
Distance / Streckenlänge [Km] : 11.5 Km
Time in seconds / Fahrzeit in Sekunden [sec] : 2353 = 39 min 13 sec = 39:13
Weight rider / Gewicht Fahrer [kg] : 69 kg
Weight bicycle, clothes etc. / Gewicht Fahrrad [kg] : 8 kg

Grade / mittlere Seigung : 10.3 %
Average speed / mittlere Geschwindigkeit : 17.5 Km/h
Total weight / Gesamtgewicht : 77.0 kg

Power : 430.6 Watt
Power / kg : 6.2 Watt / kg

]

Hi Halamala,

I'm having some difficulty trying to reconcile you data with mine.
I watched the end of the Angliru stage on internet (eurosport channel).
I also have here the data on the climb
http://autobus.cyclingnews.com/road/2002/vuelta02/?id=stages/15

during the climb I took the following notes

17:10:50 702 m (alt.flashed on screen)
17:11:43 leaders at -7km
17:13:10 Anton breaks away
17:16:16 6km to go
17:17 Cobo catches Anton
17:20:02 Cobo at -5k, 13s ahead of Froome
17:22:55 Altitude flash, couldn't read it, distracted
17:23:53 Cobo at -4km , 30 s lead (Froome)
17:28:29 cobo at - 3km 40s lead.
17:33:10 Cobo at 1390m
17:33:25 Cobo at 1400m
17:34:00 Cobo at -2 km
17:36:35 Cobo's lead 43s.
17:38:14 Cobo at -1km
17:39:22 Cobo at top of Angliru (1570m)
17:40:10 Cobo at finish.

The 702 m altitude at 17:10:50 is consistantwith other info
Putting together all the other info

Cobo climbed from 702 m to 1570m in 28:32
I guess the air density was about 1.05
I take 0.375 for his CdA
Crr~0.004
analyticcycling for
3.91 m/s
13% incline 3.91X3.6X0.13 -> 1829 m/s

Included 2.5% for transmission losses
and 69 + 7 kg =77kg
I get 418 watts

i.e 6.06 watts/kg.

If my data are correct the 6.2 watts you got for the whole 11.5 km would probably be an overestimate.

I will come back when I have time. Now, off for a spin

 

[BillytheKid]

This is all very interesting data, but there are variables that can't be quantified.
For instance, Cobo would have better knowledge of the climb. That can make a bigger difference than some would think. Still, it is understandable as to any doubts about his performance.

 

[halamala]

Hi Halamala,

I'm having some difficulty trying to reconcile you data with mine.
I watched the end of the Angliru stage on internet (eurosport channel).
I also have here the data on the climb
http://autobus.cyclingnews.com/road/2002/vuelta02/?id=stages/15

during the climb I took the following notes

17:10:50 702 m (alt.flashed on screen)
17:11:43 leaders at -7km
17:13:10 Anton breaks away
17:16:16 6km to go
17:17 Cobo catches Anton
17:20:02 Cobo at -5k, 13s ahead of Froome
17:22:55 Altitude flash, couldn't read it, distracted
17:23:53 Cobo at -4km , 30 s lead (Froome)
17:28:29 cobo at - 3km 40s lead.
17:33:10 Cobo at 1390m
17:33:25 Cobo at 1400m
17:34:00 Cobo at -2 km
17:36:35 Cobo's lead 43s.
17:38:14 Cobo at -1km
17:39:22 Cobo at top of Angliru (1570m)
17:40:10 Cobo at finish.

The 702 m altitude at 17:10:50 is consistantwith other info
Putting together all the other info

Cobo climbed from 702 m to 1570m in 28:32
I guess the air density was about 1.05
I take 0.375 for his CdA
Crr~0.004
analyticcycling for
3.91 m/s
13% incline 3.91X3.6X0.13 -> 1829 m/s

Included 2.5% for transmission losses
and 69 + 7 kg =77kg
I get 418 watts

i.e 6.06 watts/kg.

If my data are correct the 6.2 watts you got for the whole 11.5 km would probably be an overestimate.

I will come back when I have time. Now, off for a spin

Hi Le breton,

I'm going to do a more accurate calculation later today. It'll be interesting to compare our calculations.

 

[Fergoose]

For instance, Cobo would have better knowledge of the climb. That can make a bigger difference than some would think.

It could theoretically make a difference, however:

Interviewer: "Did you know the Angliru?"
Cobo: "No I didn’t know this climb."

Source

 

[Le breton]

during the climb I took the following notes

17:10:50 702 m (alt.flashed on screen)
17:11:43 leaders at -7km
17:13:10 Anton breaks away
17:16:16 6km to go
17:17 Cobo catches Anton
17:20:02 Cobo at -5k, 13s ahead of Froome
17:22:55 Altitude flash, couldn't read it, distracted
17:23:53 Cobo at -4km , 30 s lead (Froome)
17:28:29 cobo at - 3km 40s lead.
17:33:10 Cobo at 1390m
17:33:25 Cobo at 1400m
17:34:00 Cobo at -2 km
17:36:35 Cobo's lead 43s.
17:38:14 Cobo at -1km
17:39:22 Cobo at top of Angliru (1570m)
17:40:10 Cobo at finish.

The 702 m altitude at 17:10:50 is consistantwith other info
Putting together all the other info

Cobo climbed from 702 m to 1570m in 28:32
I guess the air density was about 1.05
I take 0.375 for his CdA
Crr~0.004
analyticcycling for
3.91 m/s
13% incline 3.91X3.6X0.13 -> 1829 m/s

Included 2.5% for transmission losses
and 69 + 7 kg =77kg
I get 418 watts

i.e 6.06 watts/kg.

If my data are correct the 6.2 watts you got for the whole 11.5 km would probably be an overestimate.

I will come back when I have time. Now, off for a spin

Some more consisteny checks.
On the website lavuelta.com we find a profile of the last 3 km. which shows the summit of Angliru to be 780m before the finish line and a drop of 13meters overthat distance. A bit hard to reconcile with the fact that Cobo only needed 48s to cover that distance.

at -3km the altitude is shown as 1261 m
at - 2km ............................... 1427m
at - 1km ............................... 1554m

On this other graph from 2002, the profile of the last 2 km of the climb looks quite different.

http://autobus.cyclingnews.com/road/2002/vuelta02/?id=stages/15

By interpolating we find that the point at 1261m altitude would be 2.65 km from the top of Angliru, which would imply that the distance between the top and the finish would only be 350 meters.

By interpolating we find that the point at 1427m altitude would be 1.65 km from the top of Angliru, which again would imply that the distance between the top and the finish would only be 350 meters.

For the last point at 1554m it's hard to tell.

Considering that Cobo was climbing at about 1800m/h, since
17:33:25 Cobo at 1400m
17:34:00 Cobo at -2 km

The - 2km GPS was at 1400 + 35/3600 times 1800 = 1417.5 meters which is located at km 11.138 of the full profile graph, Which would put the finish line at 11.138 + 2 = km 13.138, ie 238 meters beyond the top of the climb.

So both profiles have problems and we don't know the real profile over the last km of the climb.

However we get a consistent picture between altitudes 702m and 1528 meters using the full profile from 2002 and the split times I noted on the internet for Cobo AND a climbing rate of 1810 to 1850 m/hr.

In the end, in view of the uncertainties on the exact profile it does ot seem possible to calculate Cobo's power output at better than +/- 0.1 watts/kg.

My final word is 6.1 +/- 0.1 watts/kg.

 

[Le breton]

Some more consisteny checks.

So both profiles have problems and we don't know the real profile over the last km of the climb.

Usually i don't go to salite.ch for precision data on climbs.
Anyway, what do they have to say on Angliru?
http://www.salite.ch/angliru.asp?Ma...liwld&dx=485&dy=330&empriseW=970&empriseH=661

It turns out that their data points are in perfect agreement with


http://autobus.cyclingnews.com/road/.../?id=stages/15

but the kilometer markers are displaced by about 100 meters.

salite.ch shows a completely different profile for the end of the climb :
it shows that between km 12 and the top (located at km 12.55 as 0.55 = (1570-1516)/0.098) there is still a 9.8% incline.

So, for once the salite.ch data might be the right one.

The distance between points at altitudes 700 and 1570m would be 6.55km, and there would be no correction to make on the calculation to take account of any flattish section before the top as there would be none.

It then becomes possible to calculate the power (Cobo's) beteween altitudes 1390 m and the summit, the distance would be 1510meters, the elevation gained 180meters, air density ~1.02
We get 394 watts ie 5.7 watts/kg for the last 6 minutes of climb at around 1500m altitude.

Altogether I think we get a fairly consistent picture of Cobo's climb of Angliru.

Basically, using his power figures it is possible to state that he would have climbed the Alpe d'Huez in the 2011 TdF about 2:30 faster than Pierre Rolland ( see many pages higher up on this thread) with a climbing time around 39:30. Faster than anybody in 2011, but far slower than the best Pantani times or even slower than the likes of L.A., Ulrich, Indurain, Zulle, Riis, Virenque ...

The question now is : is that too fast to get caught?

 

[Escarabajo]

I agree with what you said, but that was not my point though.

I was referring to the most efficient way of working up the climb. Might save some energy and at the end you could have gone faster and still have the same average power.

The equation we use is intended for steady state conditions, here:

http://www.oocities.org/mdetting/sports/cycling.html

For climbs like the Angliru it becomes a non linear equation and an unsteady equation because of the nature of the climb. The Kinetic energy factor can be represented in more heat dissipated into the atmosphere.

As for the average power with accelerations and decelerations the same as the steady pace power I agree with you. Because you already know the power to begin with. But I am sure that two subjects having everything else equal, one going at constant pace and the other one going at unsteady pace, very irregular, and having the same time at the end of the climb, I can assure you the one of them employed more power to get to the top. Even if it is as small as 5 watts. Can be bigger depending on how inefficient you go on the climb. This is true for few exceptions. For the majority of the times different to Angliru might become negligible. It is usually neglected in the calculations for that reason.

Hey I am just pointing out something that is the energy equation. It is probably a moot point for the majority of the climbs, but for the Angliru could be a factor. I know in some papers I have seen them use this kinetic energy term just because they have a powerful computer to solve with numerical methods. So it exists. Small but exists.

Read the analogy that I already stated with the cars with the fuel consumption. One of them will consume more fuel even though they arrived at the same time, but one of them stopped several times along the way while the other one kept a steady pace. Small difference but still can be accounted for.

I just finished watching the stage 17 and I am going to start calling Froome "Steady Pace Froome"

LOLOLOLOLOLOL.

 

[El Pistolero]

If I am not completly mistaken Cobo wasn't in a breakaway on that stage, was he?
So either he (and Piepoli) did 6.1 or 6.5 as he came in alongsinde Piepoli.

If I remember correctly Piepoli was leading the entire climb with Cobo mostly trying to follow his wheel and Piepoli even waiting for him a few times!

 

[Juan Pelota]

Anyone have Cobo's time up the Angliru in comparison to Heras, Simoni and Contador? The whole climb..not just the last 5k?

 

[hrotha]

If I remember correctly Piepoli was leading the entire climb with Cobo mostly trying to follow his wheel and Piepoli even waiting for him a few times!

I think it was the opposite. Recently I read (dunno where) that Cobo even got mad the few times Piepoli took a turn on the front.

 

[Anonymous]

Anyone have Cobo's time up the Angliru in comparison to Heras, Simoni and Contador? The whole climb..not just the last 5k?

video here.
http://www.dailymotion.com/video/xkvexh_vae-stage-15-final-climb_sport

you can time it, we started recording at the 11.2km to go mark, which is pretty much the start of the climb.

 

[Frosty]

I think it was the opposite. Recently I read (dunno where) that Cobo even got mad the few times Piepoli took a turn on the front.

A lot of the climb is on youtube,

eg, http://www.youtube.com/watch?v=4Dk1KC9DDBQ&feature=related

 

[Escarabajo]

Anyone have Cobo's time up the Angliru in comparison to Heras, Simoni and Contador? The whole climb..not just the last 5k?

I think it is already in this thread.

 

[RChung]

+1 for your post, except on one point, this one :
A normalized power figure is designed to indicate an equivalent psychological "cost" to maintaining a constant wattage

I do not believe it is meant to be the equivalent psychological cost but the equivalent energy expenditure the rider could have sustained if he had maintained a constant power output instead of a variable one.
Also, to give credit where credit is due, it seems to me that Andy adapted Roger "1st 4-min mile" Bannister' TRIMP method based on heart rate to the avent of powermeters and calculations based on watts.

It's not the "equivalent energy expenditure" because that would just be the number of joules. If the original quote had used the word "physiological" rather than "psychological" then it would have been pretty close.

The Bannister who developed TRIMP was Eric W. Bannister, the physiologist, not Roger Bannister, the first 4-minute mile runner. Sadly, Eric Bannister passed away last year.

 

[Le breton]

It's not the "equivalent energy expenditure" because that would just be the number of joules. If the original quote had used the word "physiological" rather than "psychological" then it would have been pretty close.

The Bannister who developed TRIMP was Eric W. Bannister, the physiologist, not Roger Bannister, the first 4-minute mile runner. Sadly, Eric Bannister passed away last year.

Thanks for the Bannister clarification.

On the first point you are too subtle for me.

 

[RChung]

Well, power is pretty variable, especially in races but even sometimes just when doing normal on-the-road training. That makes mean wattage a poor estimator of the overall workout or race stress. Andy developed NP as a step in a bigger picture way of quantifying stress by intensity and "exposure time" to that intensity. From a historical perspective, he actually had the idea of a PMC first, and worked backward to figure out how to model "fitness" and "freshness" and then training stress via its components of intensity and quantity. He needed NP for the intensity measure.

 

[webvan]

Just watched my recording of pena cabarga, last km : 2'30" -> 24km/h with an average slope of 11%, not bad...

 

[Winterfold]

I think it was the opposite. Recently I read (dunno where) that Cobo even got mad the few times Piepoli took a turn on the front.

I remembered it like that but watching the vid again Pippo goes to the front once they have gapped Schleck and then is out of the saddle for pretty much the whole of the rest of climb.

There was some real comedy gold in that Tour

 

[Fergoose]

Just watched my recording of pena cabarga, last km : 2'30" -> 24km/h with an average slope of 11%, not bad...

Can any meaningful calculation be made in relation to that stage given how uneventful it was to begin with and how short the sprint was at the end?