Comprehensive Climbers Ranking

[Netserk]

formigal is lightning striking twice before the climb

2013 was a normal ascent

 

[Peyresourde]

Thanks to your question @Ulle97 I have finally been able to motivate myself enough to analyze all the Giro ascensions on Etna since 2011. The reason I hesitated so long is that I knew there had never been a great performance and that they literally use a different side of the mountain every single time.
It is really annoying to have to look up different segment details every single time.

The best performance was (as expected) Contador in 2011, but only with an Index of 71. All the other ascents were Indexes in the 60s with either a late attacker or a group arrival.

I have to agree with Red Rick and hayneplane. The main reasons for the small gaps there is the early position in the race and the fact that there is often no strong favourite in the Giro (and shallow gradient + some headwind)

 

[Red Rick]

IIRC Contadors attack on Etna was like 16 minutes at 6.4W/kg, which was pretty good at that moment in time. It was before that they were really slow

 

[Peyresourde]

IIRC Contadors attack on Etna was like 16 minutes at 6.4W/kg, which was pretty good at that moment in time. It was before that they were really slow

Yes, the attack looked quite strong. But unless it is a monumental attack, I normally only analyze the whole climb and the pacing on Etna is always pedestrian.

 

[Netserk]

i think its worth segmenting climbs when it better helps you infer the capacity of the riders

but i guess its very hard to evaluate precisely how much pacing matters

 

[Peyresourde]

i think its worth segmenting climbs when it better helps you infer the capacity of the riders

but i guess its very hard to evaluate precisely how much pacing matters

I have segmented some climbs of course. As I said, unless it is a monumental attack, the Index for the whole climb is almost always higher than just for the attack. Exceptions for this are for example Vingegaard on Tourmalet in 2023 and Pogacar on Foscagno in 2024.

On first glance, Contador's attack seems to have been around 6.2 w/kg for 15+ minutes. Not that good, but it could be more as I did not check it very thoroughly.

 

[Netserk]

I have segmented some climbs of course. As I said, unless it is a monumental attack, the Index for the whole climb is almost always higher than just for the attack. Exceptions for this are for example Vingegaard on Tourmalet in 2023 and Pogacar on Foscagno in 2024.

On first glance, Contador's attack seems to have been around 6.2 w/kg for 15+ minutes. Not that good, but it could be more as I did not check it very thoroughly.

to analyse it properly , youd need to take all segments into account

vingegaards attack on tourmalet was a much stronger performance than if the climb had only started in la mongie

thus one needs to have some model for how much the overall pacing affected the average

 

[Peyresourde]

to analyse it properly , youd need to take all segments into account

vingegaards attack on tourmalet was a much stronger performance than if the climb had only started in la mongie

thus one needs to have some model for how much the overall pacing affected the average

Of course I take in to account that the riders have climbed already at a certain pace before. But I do cap the maximum adjustment for that at some point and the way I do it is not completely scientifc.

The uncertainty and the effort required also go way up if you start segmenting a lot, so I only do it for some famous performances. I would be happy if someone managed to create a real model, but I imagine that would be very difficult and probably only possible by relying on Strava data, which many top riders do not publish (with power).

(My own calculations are mostly based on video timing and analysis. The videos often cut away from the main group causing you to miss something, sometimes there are also no clear geographical features that let you determine the exact location in the middle of a climb etc.)

 

[Netserk]

yep i also imagine that its very difficult to do so accurately ( maybe to a degree such that its a futile exercise )

i think a starting point could be two segments , as you often have a negative split before a strong attack . in that scenario the attack is usually full throttle and a pretty good test for capacity when the full context is taken into account