Comprehensive Climbers Ranking

[Netserk]

https://www.reddit.com/r/peloton/comments/8y7n27/comment/e29z866

View: https://www.reddit.com/r/peloton/comments/8y7n27/comment/e29z866/

 

[Red Rick]

https://www.reddit.com/r/peloton/comments/8y7n27/comment/e29z866

View: https://www.reddit.com/r/peloton/comments/8y7n27/comment/e29z866/

Good old Florrie

 

[Netserk]

the itzulia itt should be a good comparison for his spring level in 2016 ( although that was the last stage of the race ) as that was close to the perfect 15 minute test

or the dauphine prologue

 

[Peyresourde]

The Itzulia TT was around 7w/kg for 15 minutes, though this was not the finish yet. very good, but nowhere near 7.4+ w/kg for 20 minutes.

The numbers I have for the Dauphine prologue are so bad that they are almost unrealistic. Either I have the wrong segment, or the hairpins are messing with the estimation. I will have to check that again.

Edit: Regardless of how I look at it, the raw w/kg of the prologue are terrible (Even if I somehow underestimated them). Despite that, the gaps were even quite large, no idea what was going on that day. (Every stage race, a lot of rider seem to underperform on stage 1)

 

[Pozzovivo]

How many grand tours the last 20 years were won by a worse climber than his direct competition in that race?

Perhaps operationalized as inferior avg.index (average of top three to five) in that race.

These comes to mind as more likely than not:
Wiggins 12'
Kuss 23'
Evans 11'
Dumoulin 17'

Depending on analytic choices perhaps:
Yates 25'
Froome 17'
Contador 15'
Bernal 19'
Aru 15'
Hesjedal 12'
Roglic 23'
Pogacar 20'

 

[Red Rick]

Big difference between not having the highest and being below the top 3 average.

Problem is if you average the indexes out youre gonna favor the shorter climbs with higher variation a lot.

But the clearest non winning average should be Kuss, Contador in the 2008 Giro and probably 2012 Vuelta as well, in addition to Evans. Probably Roglic in 2020 but not in 2023, and likely Nibali in 2016, who I guess may be as low as 4th below Chaves, Valverde and Kruijswijk

 

[Krzysztof_O]

Big difference between not having the highest and being below the top 3 average.

Problem is if you average the indexes out youre gonna favor the shorter climbs with higher variation a lot.

But the clearest non winning average should be Contador in the 2008 Giro and probably 2012 Vuelta as well, in addition to Evans

If you want to average GT performance then weighted average is needed: duration or at least some power of it (between 0.5 and 1) should be a weight.

 

[Red Rick]

If you want to average GT performance then weighted average is needed: duration or at least some power of it (between 0.5 and 1) should be a weight.

Probably.

You still get funny results because you get very tactical climbs influencing results such as Sestriere 2025 or the random result everyone forgot about like Etna 2020

 

[Peyresourde]

How many grand tours the last 20 years were won by a worse climber than his direct competition in that race?

Perhaps operationalized as inferior avg.index (average of top three to five) in that race.

These comes to mind as more likely than not:
Wiggins 12'
Kuss 23'
Evans 11'
Dumoulin 17'

Depending on analytic choices perhaps:
Yates 25'
Froome 17'
Contador 15'
Bernal 19'
Aru 15'
Hesjedal 12'
Roglic 23'
Pogacar 20'

Normally it is easy to see who the best climber is in a race just by watching it. My Index is also not perfect to analyze this since riders often don't go all out or have 3 good performances but collapse later. It also weighs short climbs the same as long ones. Here is the average of the top 3 performances for the winner and the 'other best climber' in those races.
(The others have already mentioned shortcomings of this method, so use it for pure entertainment purposes.)

Tour 12: Wiggins 67.3 <--> 70.3 Froome
Vuelta 23: Kuss 80.3 <--> 86.3 Vingegaard
Tour 11: Evans 64.3 <--> 70.3 Schleck
Giro 17: Dumoulin 76 <--> 77 Quintana

Giro 25: Yates 78.7 <--> 85.3 Carapaz
Tour 17: Froome 70.3 <--> 73.3 Barguil
Giro 15: Contador 71.3 <--> 72 Landa
Tour 19: Bernal 76 <--> 77.3 Pinot

Giro 23: Roglic 80.7 <--> 77.7 Thomas
Tour 20: Pogacar 87 <--> 82.3 Roglic

I left out the Giro 12 and Vuelta 15, but Aru and Hesjedal might have been the strongest in these races.

 

[Pozzovivo]

Yeah, didn't think that question through before I asked. My bad.

But even stupid questions can provide some interesting answers, Warren Barguil in 2017? What?

 

[Red Rick]

Yeah, didn't think that question through before I asked. My bad.

But even stupid questions can provide some interesting answers, Warren Barguil in 2017? What?

Many small gaps and Izoard was IIRC the fastest climb the entire TdF.

 

[Krzysztof_O]

Probably.

You still get funny results because you get very tactical climbs influencing results such as Sestriere 2025 or the random result everyone forgot about like Etna 2020

You could select climbs that really made the difference but then again the best way to access general performance in the mountains is to look at the GC (minus TT results, echelons and crashes).

 

[Red Rick]

You could select climbs that really made the difference but then again the best way to access general performance in the mountains is to look at the GC (minus TT results, echelons and crashes).

Or maybe just the combined climbing times of all the climbs that were GC relevant. But anyway, you're always gonna have some issue or another.

 

[hayneplane]

Normally it is easy to see who the best climber is in a race just by watching it. My Index is also not perfect to analyze this since riders often don't go all out or have 3 good performances but collapse later. It also weighs short climbs the same as long ones. Here is the average of the top 3 performances for the winner and the 'other best climber' in those races.
(The others have already mentioned shortcomings of this method, so use it for pure entertainment purposes.)

Tour 12: Wiggins 67.3 <--> 70.3 Froome
Vuelta 23: Kuss 80.3 <--> 86.3 Vingegaard
Tour 11: Evans 64.3 <--> 70.3 Schleck
Giro 17: Dumoulin 76 <--> 77 Quintana

Giro 25: Yates 78.7 <--> 85.3 Carapaz
Tour 17: Froome 70.3 <--> 73.3 Barguil
Giro 15: Contador 71.3 <--> 72 Landa
Tour 19: Bernal 76 <--> 77.3 Pinot

Giro 23: Roglic 80.7 <--> 77.7 Thomas
Tour 20: Pogacar 87 <--> 82.3 Roglic

I left out the Giro 12 and Vuelta 15, but Aru and Hesjedal might have been the strongest in these races.

For tour 12 and Vuelta 23 do you have the climbing level data for the best non-teammate of the race winner.

 

[Pantani_lives]

Indurain in his hour record apparently pushed 509.5 Watts at 81 kg for 60 minutes. (Ganna beat that by like 3 kph with 50 Watts less, crazy how bad Indurain's CdA was)

I know that CdA means "coefficient of drag x area", but how do you know the CdA of an individual rider?

 

[Red Rick]

I know that CdA means "coefficient of drag x area", but how do you know the CdA of an individual rider?

Testing, or estimated based on reasonable estimates of the other variables, especially if power is known

 

[Peyresourde]

For tour 12 and Vuelta 23 do you have the climbing level data for the best non-teammate of the race winner.

For the 2012 Tour, it is Nibali with 67.3, same as Wiggins.

Vuelta 23: Ayuso 78.3, but Mas aand Landa are very close

I know that CdA means "coefficient of drag x area", but how do you know the CdA of an individual rider?

Every formula can be solved if there is just 1 unknown variable. The location and date are known, so the weather data (air pressure) can be looked up. The speed is known. If the power and weight are known, there are only 2 uncertain factors left:

CRR (Coefficent of rolling resistance) and CdA. The CRR can be estimated reasonably well, so after that you just need to solve the formula and you will get the CdA.


In reality I think it was the other way around: Indurain did not actually have a power meter for his record. But they did test his CdA in a windtunnel before and the CRR and weight was also known. So someone estimated the power that way.

And by my comment I just meant to say that Indurain's CdA must have been bad because he was 3 kph slower than Gann while pushing 50 Watts more. (CdA is by far the most important factor at high speeds)

 

[Pozzovivo]

At what speed does CDA really start to matter +/-?

Could anyone provide a practical example of say 15, 20, 25 kph for climbs?
And how much is the difference between the most aero and least aero rider on a 25 kph climb in terms of watts/kg?

 

[Berniece]

Every formula can be solved if there is just 1 unknown variable. The location and date are known, so the weather data (air pressure) can be looked up. The speed is known. If the power and weight are known, there are only 2 uncertain factors left:

But you never know the power and weight for sure, right?

 

[Peyresourde]

But you never know the power and weight for sure, right?

This comment referred to the specific example of Indurain's hour record, were all the parameters are known (approximately). For climbing efforts, I use 60kg standard weight for all calculations. I also use era-adjusted standard CdA and CRR for every rider based roughly on an average 60 kg rider.

Someone who is more aredynamic than normal has to push less power than my formula spits out, but being aero is also a skill.

At what speed does CDA really start to matter +/-?

Could anyone provide a practical example of say 15, 20, 25 kph for climbs?
And how much is the difference between the most aero and least aero rider on a 25 kph climb in terms of watts/kg?

CdA is essential for calculating the watts needed to overcome air resistance at a certain speed.
Let's use an unrealistically extreme example: Rider A with a CdA of 0.25 (mythical ultra aero climbing position), Rider B with a CdA of 0.4 (very unaero, more standing efforts).

15kph: Rider A needs 9.8W <--> 15.8 W for B
20kph: Rider A needs 23.3W <--> 37.3 W for B
25kph: Rider A needs 45.6W <--> 72.9 W for B
30kph: Rider A needs 78.8W <--> 126 W for B

This is for a climb in standard conditions. The exact watts can change depending on weather, draft and altitude.

So at 20kph the difference would already be 14 Watts = 0.23 w/kg in my example. Of course, in reality the differences in CdA are not that big (unless someone is riding out of the saddle all the time). More draft and higher altitude also lower the consequences.



More realistic example:
The CdA is just one factor in the formula and effects the watts proportionally. Double the CdA means double the watts needed to overcome air resistance. If rider A has a CdA of 0.3 and rider B of 0.3333, then Rider A only has to use 90% of the Watts to overcome air resistance at any given speed.

15kph: Rider A needs 11.8W <--> 13.1 W for B
20kph: Rider A needs 28W <--> 31.1 W for B
25kph: Rider A needs 54.7W <--> 60.8 W for B
30kph: Rider A needs 94.5W <--> 105 W for B

At 20 kph, the difference would still be relatively negligible, at 25kph it is ~ 6 Watts = 0.1 W/kg. You can also see that around 1 w/kg is needed to overcome air resistance at 25kph. The rest is divided between gravity (lion's share) and rolling resistance (and potential changes in velocity).


(All examples are for solo efforts with 0% draft. As fo the least and most aero rider, I have no idea. If you take riders with huge size/weight differences, then big diferences in CdA are possible. CdA also depends on the climbing style)