The limits of performance

Cycling fans, and fans of climbing in particular, were able to enjoy an epic Tour de France this year. While the race may have whimpered through the Pyrenees for the overall contenders, there was plenty of action in the Alps before the final torture test of Mont Ventoux.

The transformation of Bradley Wiggins from track star to a grimpeur of the highest order was one story. The other stories, perhaps to be told here later, are of the breakaway victories on major climbs, most notably on Mont Ventoux itself. The major story, though, was the incredible climbing performance of Alberto Contador.

While the Tour overall certainly raised the bar from last year – and more on this below – it was the ride by Contador on the relatively short climb to Verbier that sent commentators and pundits into a frenzy of analysis and, ultimately, finger pointing with the thinly-veiled accusation that Contador was doping.

Contador’s performance has been analyzed in a number of places, in perhaps the most detail for general readers by the very-capable chaps at the Science of Sport blog.

Their analysis relies as its starting point on the work of French sports scientist Antoine Vayer, who calculated Contador’s performance on this 20 minute climb as a power output of 490 watts (normalized for a 78 kg rider), or an absolute output of 440 watts – or 7 watts/kg for a 62 kg Contador (Michele Ferrari published figures that put Contador at a power output of 6.78 w/kg, or 420 watts).

What raised the red flags, aside from Vayer’s suggestion that Contador had an impossibly high VO2 max, is that a 6.8-7 w/kg climbing performance over a long climb is a rarity, reminiscent of Marco Pantani in the 90s – and we all know that his natural gifts were almost certainly EPO augmented.

Many were quick to point out that comparing Verbier to longer climbs was flawed analysis.

“Firstly, if you do a direct comparison with VAM from 2009 to 1996, and you’re comparing Hautacam or Alpe d’Huez with Verbier, well that’s just not a fair comparison,” Jonathan Vaughters told Cycling News. “Alpe d’Huez is about a 40-minute climb while Verbier is around half of that so the biggest error in the calculation that he’s making is that the component of anaerobic work done in a 20-minute period is 30 to 40 percent larger than the component of anaerobic work done in a 40-minute period. It’s flawed from a scientific perspective.”

Quite, and it is worth noting that, according to Cyclismag, there were 27 riders clocking over 410 watts on the climb. The Science of Sport chaps suggested that the length of the climb would not make a massive difference, however. Their figures was that wattages would be around 3% lower. This would put Contador’s equalized wattage at around 475 watts, which is what Pantani produced in 1995 when he ascended Alpe d’Huez in just under 37 minutes – the record time by most calculations.

Interestingly, Carlos Sastre – the winner of the stage on the Alpe in 2008 – produced 6.5% more wattage on Verbier than on Alpe d’Huez (where he produced 430 watts equalized for time of 39’30”). If Contador were to produce 6.5% less wattage on a 40-minute climb he would be producing just shy of 460 watts. This would be more than Lance Armstrong in 2001 and probably put Contador’s climbing time on the Alpe at close to Jan Ullrich’s in 1997, which was 37’40”.

So what are the limits of performance? Aside from the difficulties in comparing different climbs under different racing conditions, there are also other factors to consider. In particular, the relentless pursuit of technological advantage must have surely produced some gains for climbing.

“In the mid-90s bikes were 9kg [19.8 lbs], you had these heavy Carnac shoes that Pantani used to wear, and the clothing held more sweat, so four kilograms per watt for a rider like Pantani who only weighed 55 kg is around seven percent, which would have had a massive effect on his ascent velocity,” Vaughters said in his Cycling News interview. “Then there are the tiny details like wheels being more aerodynamic, ceramic bearings, and so on…”

Vaughters’ point is interesting and his Garmin team has done a lot of work on looking for small gains through technology. Wired Magazine ran an interesting profile on the technology Armstrong used in 2004. Still, bike weights continued to get cited with confusing variations. Armstrong, for example, told the Guardian this year that, “1999, my climbing bike weighed 21 pounds; 2005 – 14 pounds.” John Wilcockson reported in 1999 that Armstrong’s bike weighed “around 17 pounds, light enough so as not to require radical changes as the Tour moved into the mountains”; in 2005, if his bike was 14 pounds it would have been under the UCI’s weight limit of 6.8 kg or 15 pounds. Interestingly, way back in 1978, Lucien Van Impe’s special climbing bike in the 1976 Tour was reported to weigh 17.6 pounds (although his regular bikes were 2 pounds heavier). A more transparent discussion is clearly needed.

Cycling fans are hyper-sensitive to what might be construed as ‘impossible’ rides. For too long, fans have been frustrated at the out-of-proportion gains possible through doping and are determined not to be fooled by huge jumps in performance, which started in the 90s with the introduction of EPO. Many still remain suspicious as to the gains Lance Armstrong was able to achieve in his reign from 1999-2005.

What makes many uncomfortable about the 2009 Tour is that the average watts of the winner, calculated using the major climbs, are back at the level of ‘suspicious’ Tours, such as in 2004. According to, Armstrong in 2004 averaged 440 watts, which was Contador’s average this year. This figure is substantially above Sastre in 2008 (415 watts), Armstrong in 2005 (425 watts) and even Contador himself in 2007 (420 watts). Even Andy Schleck in second this year, with 435 watts, posted a ride equalling the greats. In the history of the Tour, only Miguel Indurain, Bjarne Riis and Pantani have gone higher, with 445 watts on average in 1995, 1996 and 1998 respectively.

This is a far cry from prior to 1994 when there were no riders at all producing over 410 watts. In 1987 on Alpe d’Huez, Lucho Herrera was at 395 watts, with Laurent Fignon and Pedro Delgado at 390 watts in 1989 – around 5.6 watts/kg. Greg LeMond was averaging around these levels on his last two Tour wins. The Science of Sport blog shows a similar historical analysis to with these figures. In 20 years, then, performances peaked in the mid 90s before falling back, but Contador is still putting out around 12% more power than LeMond to win the Tour.

Again, what are the limits to performance, and how much should these increase over time? A rider widely regarded as being pas de dopage, Bradley Wiggins, had an average output of 425 watts (according to again), which was equivalent to Armstrong in 2005 and probably would have won him the Tour against Contador in 2007. Clearly, something has changed.

Your author looks forward to further analysis from the experts, hopefully comparing short climbs to long climbs and on the benefits of greater efficiencies from riders and the possible gains from advances in equipment technology. Cycling has a variable that other sports like track and field do not – the bike. Perhaps comparisons to early generations of riders are no longer valid, except for historical interest. Doping has undoubtedly produced anomalies, but are clean riders now pushing the envelope through greater efficiencies to achieve higher wattage outputs on the road?

A contentious area of debate, with no easy conclusions. Fans have every reason to remain wary of the mountain performances that they have seen this year. At what point, though, should we start applauding the true pushing of the limits of performance?

Wiggs 1
Lean, clean, and mean?

7 thoughts on “The limits of performance

  1. So many things have changed in the last 20 years. Doping obviously. Equipment as well has improved dramatically. The other factor is the evolution of what it means to be riding “clean”. The modern fan must now factor in riders on TUEs (e.g. for exercise induced asthma medications), better understanding of altitude training, the use of oxygen tents, liberalized caffeine use and the continued emergence of non-banned products that are ahead of the anti-doping regulations (e.g. viagra use). When you factor all this in, it might just be possible for gifted athletes to compete “clean” and achieve world beating performances.

  2. I find suspicion without a mechanism to test it rather useless. That’s why, though I laud his general effort to clean up the sport, I thought Greg LeMond’s attack on Contador in LeMonde was in very poor taste. There was the implicit belief that LeMond completely understands the science, which I don’t think is true, and further that the current tests are insufficient.

    That may well be. Perhaps LeMond has the right science, and the tests are in fact insufficient, but in the big picture the current tests are the current tests. If Contador didn’t fail any, we should be looking at his training methods and propagating them, rather than shaking down his results to find the dope.

    If we’re simply going to convict our winners of doping, with the assumption that winning requires doping, then our sport is already dead.

    If however, we work to improve the tests, while maintaining belief in our current methods and our biggest stars, then we can continue to enjoy bike racing.

    That’s what I’m going to do.

  3. I must say I share similar concerns about LeMond. He was also famous in his day for eating ice cream during stage races (albeit probably not whilst in the saddle…) When accused of frivolity by his more abstemious, often French colleagues, he would say, more or less, “calories are calories, wherever they come from.” Well perhaps so, but different foods are digested and used in different ways, which differ in their metabolic effects and therefore suitability as fuels for endurance events.

    I have not the slightest doubt that LeMond’s heart is in the right place and he has certainly been treated very shabbily by the likes of Armstrong/ Trek and Landis’ lawyer, but I’m not completely sure that, when it comes to the science, he fully understands what he’s talking about.

  4. When looking at climbers in the start of the 90s, you can see that their body postures are diffrent then the riders of 2K. Today thet ride more steady, while then they moved alot, creating eneficency in power transformation. I think today riders use more scietific training.

Comments are closed.