Comprehensive Climbers Ranking

[Pozzovivo]

This post has gotten much better peer review than most academic papers I've been reading recently

You must be reviewer #2 on my latest paper.

 

[Peyresourde]

Over the last few months, I have continued to work on my database and done a few changes. Most of them are just 1 Index points here or there. There have also been a few new performances. Today I have edited all 3 rankings in the OP.


Major changes:
- After MSR, I redid all Poggio and Cipressa ascents. I have removed all Poggio mentions (uncertainty too high) and replaced last year's Cipressa ascent with the 2026 one in the 100+ ranking

- After the renewed focus on motodrafting I also rewatched LBL 24+25 and realized that Pogacar got a bit of motodraft on Redoute in both editions. I lowered the watts slightly and both ascents are now only 99 and thus fell out of the 100+ ranking

- Indurain is moved down a tier (no major changes, but he lost 1 Index point here or there, resulting in a lower overall number)

- Lipowitz added to the climber ranking. Seixas does not have enough high level performances yet to be added.

- I will distinguish between classic/hill and mountain efforts from now on. As I do a climbers ranking, I have limited every rider to a max of 2 hill type efforts (out of 10 total climbs) to calculate their ranking. Riders like Jalabert and Porte have dropped a bit in the rankings as a result.


Some more minor changes for those who are interested:

Spoiler

- Alpe d'Huez: W/kg reduced by 0.01 for most ascents
- Ax 3 Domaines: W/kg increased after reworking the segment (Froome now 80, Laiseka in 2001: 88)
- Angliru: There was a slight mistake with the adjustment and segement in some years. Recent editions like 2023 and 2025 have slightly lower Indexes now
- I redid the Tourmalet calcs from both sides, especially the eastern one. Also Sierra Nevada.
- Lowered the watts for the Peyragudes TT slightly after looking at some more strava files

- Avoriaz: I studied all the Avoriaz ascents more closely, as there have been some impressive performances. They used a slightly different segment each time and video is not always available. Ugrumov almost cracked the 100 mark there in 1994.
Hinault in 1979 climbed faster than Indurain in 1994 (slightly different side)!! This was by far his best ever climbing performance, although the exact time is not 100% certain

 

[Rou]

How are Pogacar's performances on La Redoute 99?
To me this is quite an overestimation.

 

[BubbaChuck]

How would you rate Pog in flanders,i know its hard. Motorbike.

 

[AmRacer]

How would you rate Pog in flanders,i know its hard. Motorbike.

Remember Pidcock at 2 min

View: https://m.youtube.com/watch?v=HU6DYLGuRVs

 

[Peyresourde]

How are Pogacar's performances on La Redoute 99?
To me this is quite an overestimation.

Do you hav any specific reason?

In 2025 Pogacar did 2350 VAM into a headwind after 220k of racing. (8.59 w/kg for 3:59)

Ben Healy also mentioned in a podcast that he did almost exactly 8 w/kg that day. I have him at 7.99 w/kg which fits perfectly. Healy lost 11-12 seconds and drafted much more. This is a world of difference for a 4 minute climb.
So I am quite confident in my numbers, although it is always less certain for short climbs.

How would you rate Pog in flanders,i know its hard. Motorbike.

The climbs in Flanders are not suited to this kind of analysis (too short and cobbles create extra difficulties in the calculation).

This year, Pogacar was slightly better on the 2nd and 3rd Kwaremont (cobbles only part) than in 2025 and slightly slower on most of the other climbs. The difference was that he pushed much harder and longer on the flats.
Overall, the performances seem comparable, though I would perhaps give the edge to this year.


Just the second passage of the Oude Kwaremont (wind was slightly worse there this year). Keep in mind that the exact watts are very uncertain:

Tadej Pogacar | 87 (+9): 8.89 W/kg for 2:35 on 2nd Kwaremont (Flanders 2026)
Tadej Pogacar | 82 (+7): 8.81 W/kg for 2:35 on 2nd Kwaremont (Flanders 2025)

 

[canina82]

Do you hav any specific reason?

In 2025 Pogacar did 2350 VAM into a headwind after 220k of racing. (8.59 w/kg for 3:59)

Ben Healy also mentioned in a podcast that he did almost exactly 8 w/kg that day. I have him at 7.99 w/kg which fits perfectly. Healy lost 11-12 seconds and drafted much more. This is a world of difference for a 4 minute climb.
So I am quite confident in my numbers, although it is always less certain for short climbs.


The climbs in Flanders are not suited to this kind of analysis (too short and cobbles create extra difficulties in the calculation).

This year, Pogacar was slightly better on the 2nd and 3rd Kwaremont (cobbles only part) than in 2025 and slightly slower on most of the other climbs. The difference was that he pushed much harder and longer on the flats.
Overall, the performances seem comparable, though I would perhaps give the edge to this year.


Just the second passage of the Oude Kwaremont (wind was slightly worse there this year). Keep in mind that the exact watts are very uncertain:

Tadej Pogacar | 87 (+9): 8.89 W/kg for 2:35 on 2nd Kwaremont (Flanders 2026)
Tadej Pogacar | 82 (+7): 8.81 W/kg for 2:35 on 2nd Kwaremont (Flanders 2025)

I understand what Rou is saying. I'm not questioning your wonderful job but man, Pogacar was "cruising" on La Redoute last year.

 

[Peyresourde]

I understand what Rou is saying. I'm not questioning your wonderful job but man, Pogacar was "cruising" on La Redoute last year.

Oh, I get what you mean. But with Pogacar it is sometimes difficult to see how hard he is trying. This year on the 2nd Kwaremont I also thought he was not attacking seriously, but then he was 12 seconds slower on the last ascent when he clearly tried hard.

He also did not have to attack last year (no big watt spike), so he could pace more regularly. This allowed him to push big watts without going to the absolute limit.
The riders also seem to be able to push better watts on Redoute compared to e.g. Mur de Huy (maybe to steep?).

 

[Red Rick]

Oh, I get what you mean. But with Pogacar it is sometimes difficult to see how hard he is trying. This year on the 2nd Kwaremont I also thought he was not attacking seriously, but then he was 12 seconds slower on the last ascent when he clearly tried hard.

He also did not have to attack last year (no big watt spike), so he could pace more regularly. This allowed him to push big watts without going to the absolute limit.
The riders also seem to be able to push better watts on Redoute compared to e.g. Mur de Huy (maybe to steep?).

Mur de Huy has very suboptimal pacing

 

[BubbaChuck]

Could you look into Mercks numbers.Like his hour record,its fierce debate outhere.

 

[Zoetemelk-fan]

1987 | Jean-Francois Bernard | 90 (-5): 6.28 W/kg for 58:30 on Ventoux ITT

Just bumped over this thread.
This one really hit hard, realizing how fast time flies. One of those moments "yes, ofcourse I remember" your mind keeping telling you that was just maybe a couple of decades ago. I better keep getting tight on my bike!

A diligent attempt at comparison, which generally deserves high praise, I must say.

However, and now I have only read the OP and nothing else, so I don't know how subject have been dissected.

But small adjustments in the numbers in relation to small differences in assumptions do not necessarily delivers the most valid result, if you completely omit several essential differences now and then.

- road surfaces. There is a difference between gravel and then perfectly freshly laid asphalt on climbs, where this has only been done in recent years because a modern day GT passes by. On own body old asphalt vs fresh high tech modern asphalt layer : Lac d'Emosson 1998 vs. 2019. Really bad asphalt with potholes at first attempt, while 2019 felt like a walkthrough for last k, even though 15kgs heavier me.

- in that relation: resistance in the form of rolling resistance for tires - and frictional resistance in bearings. I remember the latter as a completely different world, jumping from my 1994 MBK Columbus aluminum bike to a test ride on a 2014 Canyon bike with Ceramispeed frictionless bearings.

And then the chain. Again a completely different value. Could Coppi have climbed Sestriere in 1952 with the same 1999 Lance Hautacam invention of 115rpm "whisk wire" cadence, which subsequently became standard, what was common with chopping style disappeared almost completely from the mid-00s and is never seen now. You can spin up to infinity today, without the mechanics giving you much resistance. I can do that on my current primary carbon bike, but not on my 1994 Colombus alu bike, here the resistance torque is clearly detected

- wind and weather are complex. Wind is not just wind. In a steady gale wind of 20m/s like for example. on my cycling holidays in Fuerteventura, you can nage yourself out of it gear-wise and find a peak. At home in fluctuating wind and pmseless gust, is like a snakebite every time, draining you of strength. Even within the same mountain range, it can be wildly different how these winds behave, shaped by the immediate surrounding terrain.

- or humid weather is not just humid weather. Huge difference in pressure and temperature and the extra resistance in terms of watts it gives.

And how much should you use as a correction factor, +3 or -2 per parameter. It becomes difficult. And with data from ancient times, a job for Dr. Watson.

I do not intend to discredit the work.
On the contrary.
It is the subject of a gigantic passion for cycling, and I love it to the bone

 

[BahamontesPhantom]

I get the criteria but can I ask how much info did you get on climbers performances of the 50s and 60s?

 

[Peyresourde]

I get the criteria but can I ask how much info did you get on climbers performances of the 50s and 60s?

Not much. The important ones are in the OP. Alpe d'Huez 1952, Mont Ventoux 1958 and Puy de Dome 1959 are the more reliable/impressive ones, though there are of course a lot of uncertain factors. I also have some ascent times of Muro di Sormano.

All I can say is that Bahamontes and Gaul were probably quite impressive (at least in TTs).

Just bumped over this thread.
This one really hit hard, realizing how fast time flies. One of those moments "yes, ofcourse I remember" your mind keeping telling you that was just maybe a couple of decades ago. I better keep getting tight on my bike!

A diligent attempt at comparison, which generally deserves high praise, I must say.

However, and now I have only read the OP and nothing else, so I don't know how subject have been dissected.

But small adjustments in the numbers in relation to small differences in assumptions do not necessarily give me the most valid result, if you completely omit several essential differences now and then.

- road surfaces. There is a difference between gravel and then perfectly freshly laid asphalt on climbs, where this has only been done in recent years because a modern day GT passes by. On own body old asphalt vs fresh high tech modern asphalt layer : Lac d'Emosson 1998 vs. 2019. Really bad asphalt with potholes at first attempt, while 2019 felt like a walkthrough for last k, even though 15kgs heavier me.

- in that relation: resistance in the form of rolling resistance for tires - and frictional resistance in bearings. I remember the latter as a completely different world, jumping from my 1994 MBK Columbus aluminum bike to a test ride on a 2014 Canyon bike with Ceramispeed frictionless bearings.

And then the chain. Again a completely different value. Could Coppi have climbed Sestriere in 1952 with the same 1999 Lance Hautacam invention of 115rpm whisk wire, which subsequently became standard, what was common with chopping style disappeared almost completely from the mid-00s and is never seen now. You can spin up to infinity today, without the mechanics giving you much resistance. I can do that on my current primary carbon bike, but not on my 1994 Colombus alu bike, here the resistance torque is clearly detected

- wind and weather are complex. Wind is not just wind. In a steady gale wind of 20m/s like for example. on my cycling holidays in Fuerteventura, you can nage yourself out of it gear-wise and find a peak. At home in fluctuating wind and pmseless gust, is like a snakebite every time, draining you of strength. Even within the same mountain range, it can be wildly different how these winds behave, shaped by the immediate surrounding terrain.

- or humid weather is not just humid weather. Huge difference in pressure and temperature and the extra resistance in terms of watts it gives.

And how much should you use as a correction factor, +3 or -2 per parameter. It becomes difficult. And with data from ancient times, a job for Dr. Watson.

I do not intend to discredit the work.
On the contrary.
It is the subject of a gigantic passion for cycling, and I love it to the bone.

Thanks for your detailed response.

I do era-adjust for all the parameters you mention (rolling resistance, drivetrain efficiency, equipment weight etc.). Of course, this is a bit of guesswork and very generalized. And if there is no good video, I can't know the exact circumstances like road surface and wind.
All these factors are included in the w/kg figure directly. The point adjustments for the index are seperate and function the same for any era (see OP)

The reason why climbs are used for this type of analysis is that often ~ 80-90 % of the watts are used to overcome gravity (depending on the steepness) and for that only the weight matters and none of the more uncertain factors you mentioned.
So all of the rest (rolling resistance, air resistance, friction and drivechain losses) does matter, but not enough to totally invalidate any calculations even if some assunptions I made are wrong for sure.

Wind is indeed very tricky and can vary locally in the mountains, so there is absolutely no guarantee that I always got it right, especially for older performances.
For air density I just use a generalized formula and humidiy, air pressure and temperature play a role there, but it is not that important at climbing speeds.

 

[AlpRaid]

Not much. The important ones are in the OP. Alpe d'Huez 1952, Mont Ventoux 1958 and Puy de Dome 1959 are the more reliable/impressive ones, though there are of course a lot of uncertain factors. I also have some ascent times of Muro di Sormano.

All I can say is that Bahamontes and Gaul were probably quite impressive (at least in TTs).


Thanks for your detailed response.

I do era-adjust for all the parameters you mention (rolling resistance, drivetrain efficiency, equipment weight etc.). Of course, this is a bit of guesswork and very generalized. And if there is no good video, I can't know the exact circumstances like road surface and wind.
All these factors are included in the w/kg figure directly. The point adjustments for the index are seperate and function the same for any era (see OP)

The reason why climbs are used for this type of analysis is that often ~ 80-90 % of the watts are used to overcome gravity (depending on the steepness) and for that only the weight matters and none of the more uncertain factors you mentioned.
So all of the rest (rolling resistance, air resistance, friction and drivechain losses) does matter, but not enough to totally invalidate any calculations even if some assunptions I made are wrong for sure.

Wind is indeed very tricky and can vary locally in the mountains, so there is absolutely no guarantee that I always got it right, especially for older performances.
For air density I just use a generalized formula and humidiy, air pressure and temperature play a role there, but it is not that important at climbing speeds.

what about gear ratios that are now so different than back then?

 

[Zoetemelk-fan]

what about gear ratios that are now so different than back then?

That was what I referred to with "whisk wire" fast pedalling introduced by Lance, while the "chopping style" low cadance had higher peak power at harder peak load on knees and thighs.
And now up to 13 chainrings at the back, where max 8-9 35 years ago.
In addition to that with resistance torque qua huge development in frictionless bearings and chain.

Would be interesting to see what Fausto could have done with modern softpedalling high cadence style and a modern frictionless low weight bike

 

[AlpRaid]

That was what I referred to with "whisk wire" fast pedalling introduced by Lance, while the "chopping style" low cadance had higher peak power at harder peak load on knees and thighs.
And now up to 13 chainrings at the back, where max 8-9 35 years ago.
In addition to that with resistance torque qua huge development in frictionless bearings and chain.

Would be interesting to see what Fausto could have done with modern softpedalling high cadence style and a modern frictionless low weight bike

please do not refer to armstrong -- it is meaningless. the spinning style was largely due to the fact that he had to focus on his "enhanced" ability to process more O2.

best cadence climbing is lower and LeMond refers to this on The Roadman (?) podcast.

But easier gears going up Tre Cime, let's say, allows for a completely different performance than when you see Gimondi, Merckx and Fuente contorted over their machines and grinding up in 1973.

 

[Zoetemelk-fan]

please do not refer to armstrong -- it is meaningless. the spinning style was largely due to the fact that he had to focus on his "enhanced" ability to process more O2.

No issue skipping LA reference if that is used to redirect my point. However the "chopping style" has been completely gonr for plus 15 years in the peleton.

best cadence climbing is lower and LeMond refers to this on The Roadman (?) podcast.

I watched The Roadman Podcast Greg Lemond interview in full lenght as soon as it landed and surely noted it. Nomatter how much I admire and respect Greg this is simply not true

But easier gears going up Tre Cime, let's say, allows for a completely different performance than when you see Gimondi, Merckx and Fuente contorted over their machines and grinding up in 1973.

Higher cadance at steeper sections is clearly an evolvement over the last 30 years and tve pattern is clear. And taking my own 65rpm style compared with now 85-95rpm (70 when extremely steep, but maybe just 50 rpm in the 80ies on those muritos) it is clear to me that making a round thread, disteibuting power is much, MUCH more efficient.

 

[BubbaChuck]

Do you have any estimations for PR. What was Pog pushing in comp to VDP. It looked VDP was stronger.

 

[vappa]

How does today's Yates's performance compare to those of Jonas and Seixas?

 

[Red Rick]

How does today's Yates's performance compare to those of Jonas and Seixas?

If you mean W/kg wise the lazy answer is just blindly looking at watts2win and they have 6.8 for 15 minutes.

If we're talking about gaps, then 45s on a 15 minute climb is really good, regardless of the competition, but with a cat 2 directly before the climb and the climb having a big ramp right at the start makes it a lot less crazy.

But really just given Adam Yates' history there is no reason to assume he could compete with Seixas let alone Vingegaard even if the watts calculators *** out 7.5 W/kg here.

 

[Peyresourde]

what about gear ratios that are now so different than back then?

Gear ratios are not part of my consideration. It is certainly an advantage to have smaller gears available, especially for consecutive climbing where you put less strain on your muscles. But this is too minor/uncertain to include in my calculation.

Do you have any estimations for PR. What was Pog pushing in comp to VDP. It looked VDP was stronger.

It is not possible to estimate watts for Roubaix. The reason these estimations can only really be done for climbs is that even minor variations in CdA totally change the watts a rider needs to push at high speeds.

MvdP and Pogacar were certainly the 2 strongest riders. My guess is that without any mechanicals it would have been a two-up sprint in the velodrome with van der Poel winning (They certainly would not have been able to drop each other).

I can't tell who was 'stronger', but one thing I know is that generally you arrive more tired if you positive split an effort (push harder in the beginning) even if you push the same watts overall. In this context, Pogacar having to push hard to come back over 100k from the finish probably cost him more than if he had made a similar effort 30k later. (But he was still more lucky than van der Poel who had his mech at the worst possible time)

How does today's Yates's performance compare to those of Jonas and Seixas?

I did not have much time this week and have not even watched the race yet. If you look at the last ~5 years, Adam Yates is probably at least the 4th best climber for fresher effort/ 1 week races (after Pogacar, Vingeggaard, Roglic).

He seems to be in good shape again now. It was probably an Index in the high 70s today, worse than Vingegaard on Pal and much worse than Seixas.
I will probably post the watts tomorrow together with Monte Trega on stage 5, which I think Vingeggaard has done in 2023.