Vuelta a España climbing speeds: how should we interpret the data?
This year’s Vuelta a España saw an outcome the bookies didn’t anticipate several weeks beforehand; Alberto Contador (Tinkoff Saxo) went from being a rider originally not even listed in the betting odds to one who took the final red jersey of overall winner.
The Spaniard bounced back from a fractured leg to prove himself as the strongest in the Vuelta a España, beating closest rival Chris Froome (Sky), another who suffered fractures in the Tour, by one minute ten seconds.
Former race winner Alejandro Valverde (Movistar) was third, one minute 50 seconds off, while Joaquim Rodriguez (Katusha) finished three minutes 25 seconds behind, consequently missing out on the podium by one slot.
The race ended some time ago but the analysis and reflections continue. As was the case after the Tour de France, CyclingTips has spoken to Dr. Mike Puchowicz, Ross Tucker and the Finnish Twitter analyst @ammattipyöräily, gathering their thoughts.
They are amongst the best-known analysts of power data and climbing speeds. While the topic of making such calculations and deriving conclusions is controversial for some, others regard it as a useful tool in analysing the sport and in forming opinions about the performances involved.
Puchowicz is a sports medicine physician from Arizona working in collegiate sports. He is actively pursuing research in performance physiology and consults on anti-doping programs internationally.
Ross Tucker is an exercise physiologist who specialised in fatigue and the limits to performance, and currently researches and lectures at the University of Cape Town, South Africa. He also runs The Science of Sport website, where he has been analysing performances in various sports including the Tour de France since 2007.
@ammattipyöräily has been following pro-cycling since 1994, and has collected climbing times and estimated power outputs since 2008. He told CyclingTips that he chooses not to give his full identity ‘for personal reasons,’ but he is regarded as perhaps the most accurate in terms of extrapolating power output data from times on climbs.
Each agreed to answer questions on the data gleaned from this year’s Vuelta. As can be seen, their opinions sync on some topics and vary on others; read on for their take on the race.
CyclingTips: What are your general thoughts on the climbing speeds of the Vuelta?
Ross Tucker: I think we saw more or less what we would’ve seen at the Tour, in terms of the level that Contador and Froome were at.
It means that we now have some indication, at least, that barring crashes, we would have watched a three-man race at power output levels that I have to say are not particularly reassuring, because they’re very similar to what would have been seen with known doping.
The other thing is that both Froome and Contador came off enforced layoffs. Did that help them? That’s a very optimistic way of arguing it. Going into the race, based on everything known about preparation and training, you’d have to think both would have been below their very best levels.
Maybe by week three they were performing at a better level than they might have had they come in with solid training, but still, you wouldn’t choose to prepare for a race with a broken wrist or tibia. So to be at the level they were is surprising.
@ammattipyöräily: Some climbs were climbed at very high speed. We saw also very tactical riding with slow speed. For me, it was like a cat-and-mouse game.
Mike Puchowicz: Overall, the performances from Contador and Froome in the 2014 Vuelta were very fast but inconsistent with two of the three longer climbs falling below the trend of the remaining six climbs.
CyclingTips: Does the shortness of some of the climbs cause a complication and, if so, is it still possible to form a conclusion?
Ross Tucker: In terms of the shortness, that’s a consideration, certainly. The model has been made to account for that, but it’s early days yet and of course the concession must be made that it can improve in time. So the direct comparison must be carefully made. That said, there were Tour climbs in 2013 and 2014 that were the same length as Vuelta climbs. See this graph by Mike Puchowicz:
@ammattipyöräily: There were only two climbs in this year’s Vuelta which were more than 20 minutes. As many as six climbs were ridden in less than 20 minutes. Thus it’s difficult to compare this year’s performances to previous years.
Of course there was also the third climb which took more than 20 minutes: Stage 11, from Pamplona to San Miguel de Aralar. However San Miguel de Aralar wasn’t ridden full gas. It was more tactical battle on that climb; Froome was setting his own tempo and the Spanish riders were watching each other.
As a result we didn’t see super high numbers then. Thus there were small time gaps in stage 11 to San Miguel de Aralar and stage 15 to Lagos de Covadonga.
The Spanish riders were pretty impulsive and it seems they rode like man against man. In contrast Froome focused on riding his own pace. The best example was seen on La Camperona in stage 14; he seemed to be in serious difficulties in the first few hundreds meters but at the end he beat all his rivals.
Behind Contador and Froome, Valverde was at higher level at Vuelta compared to his form in Tour de France. Valverde was approximately as strong as last year’s Vuelta. Rodriguez was roughly the same level as 2013 Vuelta.
Quintana’s abandonment was unfortunate, as otherwise he would have fought for the GC victory.
I also think Contador’s team was weak but his rivals were unable to use the opportunity. Froome had the strongest team-mates in the mountain stages.
CyclingTips: How do these Vuelta climbing speeds compare to those in last year’s race?
@ammattipyöräily: Last year Chris Horner set the second fastest climbing time of all-time on Angliru (Click here for the all-time list)
Horner improved Joaquim Rodriguez’s previous record time on Peña Cabarga by 16 seconds (Click here for the all-time list)
This year’s climbs were rarely used except Lagos de Covadonga. As I said it’s difficult to compare the 2013 & 2014 performances with each other.
CyclingTips: How do they compare to Nibali in this year’s Tour de France plus Froome in last year’s Tour?
Ross Tucker: It’s difficult to separate them on the shorter climbs – in fact, if anything, Froome 2013 was at a slightly higher level than Contador in the 2014 Vuelta (with one exception). Nibali was slower on the 2014 Tour’s one shorter climb, but then he wasn’t really challenged to be faster. He was however better on the longer climbs, which is the big question mark.
Regarding this year’s Tour, Froome and Contador, as I’m sure you know, produced performances that were at least comparable to Nibali at the Tour, on the shorter climbs. This post by Mike P has the numbers:
So my overall assessment – Contador and Froome in the 2014 Vuelta were in the same kind of condition Froome had in the 2013 Tour de France, at least for the shorter climbs. We will never know how they’d have fared on the Tour’s longer climbs against Nibali.
@ammattipyöräily: There were too few long climbs to calculate average power outputs. Anyway I think Contador and Froome weren’t as strong as Nibali in this year’s Tour and Froome in 2013 Tour.
Contador and Froome abandoned the Tour de France due to injuries. Obviously they weren’t in the best condition in Vuelta a España. The Tinkoff-Saxo owner Oleg Tinkov has said that Contador would have won the Tour de France despite Nibali leading Contador by 2.34 before stage ten.
In August I thought Tinkov was joking but I had to change my opinion after Vuelta a España. Contador was much stronger than I expected at Vuelta.
Mike Puchowicz: Compared to last year, Contador and Froome appeared to be on the same level as Horner in the 2013 Vuelta. On the other hand, Contador and Froome were both likely better than Nibali’s 2013 Vuelta performance.
When looking at data across 2013 and 2014 from the Tour de France and Vuelta the most interesting trend is Froome’s remarkable consistency. Despite very different build ups, and very different races, Froome’s performances are consistent with him pacing himself according to an established power duration curve.
CyclingTips: How do you feel about these figures – is there cause for concern?
Ross Tucker: I’ll say first that the physiological implications of these performances are extra-ordinary, but not superhuman, if that makes sense. My approach was always to assess what kind of physiology would be required, and to ride at these levels is, as you’d expect, on the upper limit of what is physiologically plausible (it should be, right? Because you’re watching the three best cyclists in the world – if they weren’t physiologically exceptional there’d be something wrong with the sport).
I guess what is of concern, and I’m repeating what I have said before, is that the entire picture is not reassuring. We know that there are isolated cases of guys in the middle of the pack are doping still. Is it conceivable that the guys in the middle are doping but the guys at the front are not? Are there “lone wolf” dopers, or are these cases symptoms of a bigger problem? I’m leaning to the latter.
We recognise that the attitudes of team leaders and directors and some riders have not changed or are changing slowly. The lack of due diligence and the claims of naivety from teams – you yourself wrote one of the most comprehensive articles on this re Jonathan Tiernan Locke – is a concern too.
There’s enough smoke now that I don’t think it’s possible to simply look at numbers and say “The sport is clean”, or “the sport is dirty”. People are doing that – I saw LeMond interviewed recently, and his argument is that the climbing speeds are now slower and therefore plausible. Well, only kind of Greg. Much like a politician running for office can twist the facts and figures to suit their leaning, I think people in cycling can make one of two arguments for cycling’s bill of health. There’s a PR component to the sport’s current situation that I distrust much more than the actual scientific/doping aspect.
So when you add in the ‘smoke’ – and I’m not a fan of allegation and circumstantial evidence, but at some point, it kind of demands attention! – then the direction you should lean in starts to become a little difficult to avoid!
The power outputs they’re achieving are high enough that we can say with certainty that they’d have been competitive in the period prior to the biological passport, post-EPO testing (2002 onwards).
However, it’s still slower than it was in the 1990s where all the authorities had were haematocrit ‘limits”, and the record times for most of the climbs were set. This is why the direct comparison of say, Alpe d’Huez, the Tourmalet, Luz Ardiden, etc – the iconic climbs of the Tour – will provide the best indications.
@ammattipyöräily: I think it’s useless to look power numbers on short climbs. They are always very high.
We saw the most convincing performances in stage 16 and 20. Contador and Froome climbed La Farrapona, the mountain top finish in stage 16, at 6.3 – 6.4 W/kg for 18 minutes. There was a tailwind, so the power outputs were likely a little bit lower than originally was thought. On the last climb of the Vuelta, the Puerto de Ancares, their estimated average power output was higher than 6.0 W/kg for 39 minutes. Froome was setting the pace almost all the time and Contador was on British rider’s rear wheel.
Contador and Froome’s ride on Puerto de Ancares is very comparable to Nibali’s performance on Hautcaam in stage 18 of Tour de France. Nibali climbed at 6.0-6.1 W/kg for 37 minutes and 20 seconds.
Mike Puchowicz: In terms of comparing this Vuelta to the 2002-2007 doping era baseline, the interpretation is a bit tricky. The issue is that I don’t have enough data from the short climbs from 2002-2007 to make direct comparison for all the climbs. If just the three longer climbs are considered then two out of these three climbs are well below the doping baseline. The implication from this comparison would be a lower likelihood of enhanced performance.
On the other hand, since these two slower climbs seem to be the outliers, my overall assessment is that the data is inconsistent and not specifically reassuring.