Alpe d’Huez revisited

The iconic climb of Alpe d’Huez featured again at the Tour de France this year and proved to be decisive in the outcome of the race. Le grimpeur has already discussed this climb on several occasions (see here, for example), but the publication of a new book on l’Alpe, as well as questions arising from this year’s stage suggests that further discussion is of interest.

This year’s battle on Alpe d’Huez was a thrilling one (AFP pic)

In his excellent new book, ‘The Tour is Won on the Alpe’, French journalist Jean-Paul Vespini lays to rest the long debate over the fastest ascent of the Alpe d’Huez climb at the Tour de France. Controversy has surrounded this question given that times have been recorded for different distances. For example, for the 14.5 kilometre distance, which was used for official timings since 1990, Marco Pantani holds the record of 37’35″, set in 1997, which is typically cited as the fastest ascent of the climb in numerous sources and articles.

Once the short stretch of some 700 metres from the turnoff just outside Bourg d’Oisans is omitted, however, the distance is shortened to 13.8 kilometres for the actual climb. According to Vespini, Pantani still holds the record for this shorter distance, but for the time of his 1995 stage win: 36’50″. He lists the top-ten times for this distance as follows.

1. Pantani 36’50″, 1995, 22.5 kph
2. Pantani 36’55″, 1997
3. Pantani 37’15″, 1994
4. Armstrong 37’36″, 2004 (ITT)
5. Ullrich 37’40″, 1997
6. Armstrong 38’01″, 2001
7. Indurain/Zulle 38’10″, 1995
8. Riis 38’15″, 1995
9. Virenque 38’20″, 1997
10. Landis 38’34″, 2006, 21.5 kph

These records are startling for several reasons. Firstly, that Pantani holds the top three placings. Secondly, that 70% are in the years 1994, 95 or 97. Thirdly, that no climbers from 1976 to 1992, when the Alpe was climbed 15 times at the Tour, make it onto the record books (the climb itself was not timed before 1990, but no one is suggesting that any of the earlier climbs were records). Other sub-40′ climbs from various sources, Andreas Kloden in 2006 and Iban Mayo in 2003, fall into the same era.

The difference of the mid-1990s onward is stark. According to Vespini, in 1990, the ascent was timed over 15.5 kilometres. Erik Breukink managed 43’19″, with Fabio Parra at 43’23″, Andy Hampsten at 43’58″, and Claude Criquielion at 44’00″. These are still impressive rides, with average speeds in the 21.5-21 kph range – at the same level as Landis in 2006. But the extra 1.7 kilometres from Bourg d’Oisans is dead flat, meaning that speeds approaching the climb would have been higher. If we assume, say, that Breukink was attacking the base of the climb at 35 kph (he was probably faster), he would have covered the distance in 2’54″, making his time for the climb 40’25″ (around 20.5 kph). These were the top climbers of the day, but in 1990 they were still nearly 2 minutes short of a top-ten performance.

(The excellent writers at, for which much the subsequent discussion below depends, offer a list of historical times, the fastest of which is Luis Herrera in 1987 with 41’50″. It seems that they have, unfortunately, not differentiated between the various distances used in the timings – making first-glance comparisons impossible. They do note, however, that the first time below 40 minutes was set by Gianni Bugno and Miguel Indurain in 1991 (Indurain struggled in at 58 minutes in 1989), which is likely for the 14.5 kilometre distance.)

l’Alpe saw some great climbing in 2006

As has already been discussed on le grimpeur, the record times for Alpe d’Huez are clouded with suspicion. With just quick glance at the above list, we have Landis convicted of doping in the same Tour, Virenque having confessed to doping, Riis also having confessed, Zulle also admitting to doping at ONCE, various members of the Ullrich’s Telekom squad having revealed a team doping plan at the time, various documented sources of Pantani’s likely doping, and ongoing speculation and circumstantial evidence around Armstrong’s performances. Add in suspicions about Indurain and there is – at the minimum – an aura of disbelief about these times.

That the records coincide with the EPO and blood-doping era in cycling seems to be no coincidence. It is well known that such doping methods are more effective at improving performances than the drugs, principally steroids, used in the 70s and 80s. Sub-40 minute times were not recorded until the era of these doping methods. This does not mean that all the records prior to 1994 (or 1991, Bugno and Indurain) were done pas de dopage but simply that subsequent rides have benefited from better doping products on a scale from definitely to more than likely.

To break it down in simple terms, these top-ten riders were simply putting out more power. According to, equalizing for rider weights, Landis produced 442 watts in 2006, Armstrong 450 watts in 2001, and Pantani 470 watts in 1995 – 6.3, 6.4, and 6.7 watts/kg respectively.

In comparison, in this year’s Tour, Carlos Sastre’s winning time of 39’30″ maxed out at 430 watts, with his chasers not even close to this level – Andy Schleck, Sammy Sanchez, and Alejandro Valverde at 411 watts, and Frank Schleck (the 2006 stage winner with a 40′+ time), Cadel Evans, Christian Vande Velde, Bernard Kohl, and Dennis Menchov and others all at 403 watts. All posted times over 41 minutes. By comparison, Valverde, for example, was therefore producing 6% less power than Mayo in 2003 and a staggering 15% less than Pantani tapped out in 1995. Calculating power, 411 is 5.8 and 403 watts is 5.7 watts/kg.

This discussion invites three conclusions, one old and two new. Firstly, as noted here previously, Pantani’s record for the climb may never be broken. His incredible performance is so head-and-shoulders above anything seen in recent years that it may not even be possible to replicate again, thus cementing his reputation as the fastest grimpeur in the history of cycling. Still, we might be tempted to join the somewhat fractious Federico Bahamontes, who dismissed Pantani in a recent interview in Rouleur magazine as simply being “high as a kite”.

Secondly, it puts Carlos Sastre’s 39’30″ ride at this year’s Tour into perspective. We might assume that, in this supposed new era of cycling, where rider speeds in the mountains appear to be mostly, if not uniformly, slower that a sub-40 minute ascent of l’Alpe is exceptional. Sastre appeared untouchable on the climb, and the numbers confirm it. The chasing group may have been slowed by the tactics of the Schleck brothers but, free to ride to his limit, Sastre rode at a level not far short of the top-ten list and about the same as the performances recorded in the early 90s. Cyclismag also noted that Sastre clocked sub-40′ times in 2004 and 2006, which can only lead to the conclusion that he is indeed one of the very best climbers in cycling.

This has naturally led to much speculation about Sastre, given how sensitive we are in current times over exceptional performances. Cyclismag notes that “Sastre ne semble pas fatiguer après plusieurs ascensions“, he is not tired after several climbs, showing how his three times for l’Alpe are quite similar, despite two being at the end of long road stages and one in the ITT of 2004. By comparison, Moncoutié and Goubert recorded times this year around 3-4 minutes slower than in 2004. Somewhat cryptically, Cyclismag concludes that “L’effet de la fatigue est clairement visible pour ces deux coureurs au contraire de Sastre.” Cyclismag also notes that Sastre’s ride was “du même acabit“, directly comparable to the performance by Ricco and Piepoli in the Pyrenees – now tainted by use of the third-generation EPO drug Micera. At least for now, more concrete conclusions on this debate remain elusive.

Leipheimer, Sastre, and Landis in 2006

Thirdly, and finally, we can speculate on the implications of other performances on l’Alpe. There has been much recent discussion on the end of pure climbers in cycling, the seemingly mythical capabilities of riders such as Bahamontes, Gaul, Van Impe, and Herrera to slip away from the rest of the field in the mountains. The finger has – rightly – been pointed at EPO and blood doping. “Suddenly it’s no longer an advantage to be ultra-light and lean,” former Festina coach Antoine Vayer told Procycling last year. “If a rider is heavier because they’re pumped up with anabolic steroids, as long as their blood is well oxygenated they can win mountain stages.” Others, though, have also noted that training methods for climbing have improved. The careful use of power meters has enable for focused training and the attention to incremental gains that were not possible in previous eras. “Increasingly these days we notice how the non-climbers are making big improvements [in climbing] while the natural climbers make very small gains,” the DS of team Lampre, Giuseppe Martinelli, said in the same Procycling story.

It is tempting to make some rough adjustments for historical times, assuming that such times listed by Cyclismag are for the slightly longer distance from the turnoff outside Bourg d’Oisans, 14.5 kilometres (Vespini seems to suggest that the 15.5 kilometre distance from the centre of the town is more likely), then we might drop 1 minute from the times of climbers in the 1980s (one might reasonably drop 2′ for a longer distance). This would put Herrera at 40’50″, Delgado and Fignon in 1989 at 41’15″, and Gert Jan Theunisse in 1988 at 42’50″ (and Hampsten et al, as noted above, probably around 42′). The 41’43″ and 41’45″ times posted this year by the next nine finishers, puts them solidly in this range, though – according to adjusted historical times – still slower than the great climbers of the previous era, Delgado and Herrera (for more on the 1987 Tour, see here).

It is also tempting, therefore, to conclude that the climb times from this year’s Tour are more in keeping with reasonable expectations as to how fast l’Alpe can be climbed, given historical times and the small gains that can be made through modern training. The substantially faster times of the mid-1990s and recent years do seem out-of-place, it can be reasonably argued, now that we have a snapshot of what is possible on the mountain under what many have pointed out is a much stricter testing regime and a cleaner peloton.

The bias of le grimpeur is towards the climbers of the 1980s, perhaps the last generation of the pure mountain men. “When I saw riders with fat asses climbing like airplanes, that’s when I knew,” said Luis Herrera, and indeed many riders of much larger builds have indeed defied gravity to ascend faster than le petit jardinier (the total wattage produced by Indurain in 1995 must have been astronomical). It now seems possible, though, to compare today’s climbers to those of Herrera’s era and to argue that the former are close but still not quite measuring up to the ‘greats’ of a previous era.

There is still further analysis to be done on climbing Alpe d’Huez and the conclusions offered here remain tentative at best, a combination of calculation and speculation. Pantani’s record ascent will remain clouded with controversy but is unlikely to ever be broken. But the top-ten times, we can now see, are exceptions – either the products of exceptional rides or exceptional preparations. Draw your own conclusions.

The remarkable Little Gardener in 1987 (GW pic)

6 thoughts on “Alpe d’Huez revisited

  1. maybe I missed it, but has there been any discussion of changed in road conditions or equipment? better paving could certainly make a difference.

  2. “Bernhard Kohl tested non-negative for CERA not once but twice at the Tour de France. Samples taken on July 3, two days before the race started, and July 15, the rest day in Pau, were both positive, the French Anti-Doping Agency AFLD has confirmed.” – CyclingNews, Oct 14.

    Plus ca change…

  3. pantani era semplicemente di un altro pianeta, nel 94 riprese parecchi fuggittivi, in particolare fece storia il recupero di pensec che allargò le braccia.
    mentre nel 95 riprese 1 minuto al gruppo di virenque,escartin,gotti,jalabert, e poi li staccò con estrema facilità.
    nel 97 una sua progressione impressionante gli permetteva di staccare ogni 500m prima riis,poi virenque e infine ullrich!
    simple the best!

  4. Does anyone out there have copies of live broadcast coverage of Marco Pantani’s ascents up Alpe d’Huez from 1995 and 1997?

    I am doing some research on this area and am desperately trying to find (as unedited as possible) coverage of these climbs.

    If anyone can help, I would love to hear from you. My email address is


Comments are closed.