Welcome to the Science of Sport, where we bring you the second, third, and fourth level of analysis you will not find anywhere else.

Be it doping in sport, hot topics like Caster Semenya or Oscar Pistorius, or the dehydration myth, we try to translate the science behind sports and sports performance.

Consider a donation if you like what you see here!

Did you know?
We published The Runner's Body in May 2009. With an average 4.4/5 stars on Amazon.com, it has been receiving positive reviews from runners and non-runners alike.

Available for the Kindle and also in the traditional paper back. It will make a great gift for the runners you know, and helps support our work here on The Science of Sport.

Thursday, July 23, 2009

Tour de France 2009: Time-trial analysis

Contador reigns supreme in yellow on the flat roads. Analysis of the Tour's final time-trial contests

Alberto Contador showed in the Alps that he is the best climber in the Tour (regardless of what the polka dot jersey says), and today, he underlined his status as the number 1 rider in the 2009 Tour by winning the 40.5km individual time-trial in Annecy.

For the Tour's best climber to do so well in a time-trial on what was largely a flat course means that Contador's rise to world's best stage racer has now been punctuated by a great ability against the clock, when he used to be a rider who gained time in the mountains and then minimized his losses in the time-trials.

Today, he was fastest from the start, and while Fabian Cancellara did a storming ride in the final 12km, it was not enough to overhaul the time that had been lost in the first part of the course.

Contador took the stage in 48:30, narrowly beating Cancellara (3 seconds), with the GC contenders stacked in amongst time-trial specialists further down. Brad Wiggins was the best of them in sixth, Andreas Kloden was 9th, Lance Armstrong 16th, and Andy Schleck 21st. Frank Schleck had probably the worst day of the big favourites, losing his podium place and leaving himself with a great deal of work if he is to reclaim it by Sunday's finish.

Implications for the GC

In terms of the GC, the big movers of the day were Armstrong, who climbed to 3rd, displacing Frank Schleck, who finished the day 1:04 behind him. Andy Schleck will also be pleased with his performance, because he limited his losses to Kloden and Armstrong to only 47 seconds and 15 seconds respectively.

It means that Andy's second overall is protected by 1:14 over Lance Armstrong. It seems a difficult ask for Armstrong to overhaul that deficit with only the climb of Mont Ventoux to come. though it could happen - a bad day or a good day either way and minutes can be lost. More interesting for Saturday's big stage will be the battle for third on the podium, since Armstrong is separated by only 11 seconds from Wiggins, with Kloden and Frank Schleck also within striking distance of a spot on the podium. The race for yellow may have been sewn up, but Saturday on Mont Ventoux should produce some fireworks for the remaining podium places.

Pacing analysis - contest by contest

I thought an interesting approach to the time trial would be to examine the pacing strategies employed by the GC contenders, and Fabian Cancellara, specifically to see who had the strength over the latter part of the race.

I've tabled the times for each of five segments of the 6 GC contenders and Cancellara, and shown where each was ranked for that particular segment. So below are two images, the first showing the rankings and times for the first three segments, the second showing the last two and the overall time. Briefly, the segments were: 0 to 18km (18km); 18km to 25km (7km); 25 to 28.5km (3.5km, including the day's short climb); 28.5 to 37km (8.5km) and 37 to 40.5km (3.5km).

Given that this time-trial was all about clawing back time, gaining time, and restricting time losses, it's interesting to compare riders in those mini-battles - Armstrong vs Andy and Frank Schleck, Wiggings vs Armstrong and Contador vs Cancellara (for the stage, rather than the GC!) - and see where they made up time and lost it. Bear in mind that when you look at these times, you're seeing effects of rider fatigue (when they slow down), but also of conscious decisions.

Contador vs Cancellara - tale of two halves

Contador made an incredibly fast start - he was 18 seconds up on Wiggins, and a full 37 seconds ahead of Cancellara, who had begun far more conservatively. That was switched around for segment 2, where it was Cancellara who was fastest, and Contador, perhaps by then a bit more settled, slow down somewhat, posting only the 5th best time of our seven riders. The trend for Frank Schleck was set from the outset, and he would go on to post the slowest time of our seven riders at all the time checks with the exception of the short climb, which is where the data is perhaps most interesting.

It was on this climb (segment 3) that Contador really won the day. His time here, over just 3.5km, was 18 seconds faster than the next best rider, and a full 31 seconds faster than Cancellara. It meant that Contador was now 46 seconds ahead of Cancellara, with the long gradual descent to come. Despite Cancellara's very fast finish (shown below), Contador held on for the win. In fact, having reached the high point of the course 46 seconds down, Cancellara managed to finish 43 seconds faster than Contador over the final 12.5 km, and eventually "ran out of road" to win his second Tour stage.

Armstrong vs Andy Schleck - the impact of the hill

Of interest on this climb is that the slowest time of all the riders was that of Armstrong, in 9:11. That is a full 46 seconds slower than Contador, and is where Armstrong really began to battle in his efforts to claw back Andy Schleck. Coming into the stage, Armstrong was looking for 1:29 to catch Andy.

Up to the bottom of the climb (25km) it was going well. In fact, before the climb (at the 25km mark), he was 33 seconds ahead of Andy Schleck on the stage. By the finish of the stage, that lead had been cut to 15 seconds, thanks largely to the 28 seconds Schleck was able to claw back on the short climb. Shleck ended up 1:14 ahead of Armstrong for second place overall, but it was looking much more tenuous until the climb gave the younger Shleck some time back.

Looking at the second group of segments, the profile is overall downhill, and so not surprisingly, the speeds are much higher, and Cancellara dominated in these two segments.

Brad Wiggins - a fast start that failed to continue

Brad Wiggins posted consistent times in terms of the rankings, but in terms of times, he faltered a little after the summit of the climb . His podium aspirations required that he make up a deficit of 58 seconds on Lance Armstrong. By the time they reached the summit at checkpoint 3, he'd recovered 42 of those seconds. However, over the final 12.5km downhill section, he was able to extend this by only 5 more seconds, to make up 47 seconds. That leaves him with 11 seconds for the Mont Ventoux, as explained above, and that may just prove too difficult an ask given the last two days in the Alps, where Armstrong looked to have his measure.

Frank Schleck - consistent time losses

Starting the day in third, it was always going to be tough for Frank Schleck to keep Armstrong at bay. He had a 30 second advantage, and it was gone by the first time check at 18km, where he was already 35 seconds down. From that point onwards, he "only" lost another 30 seconds, thanks in part to the climb, where he recovered 19 seconds on Armstrong, who did that climb slowest of all, as mentioned. That means he goes into the Mont Ventoux stage with 34 seconds to make up to catch Armstrong. That may just prove a bridge too far, unless he has an incredible day and is able to use the steep part of the climb and perhaps his brother's support to attack. He does of course also have to worry about Kloden and Wiggins and make sure he beats them by almost the same margin, so the chances of a podium finish seem slim.

An early predication for the podium then says that Andy Schleck will hold on, and that Lance Armstrong should have enough to hold off Wiggins and Frank Schleck for the third spot.

Looking ahead

Tomorrow is likely to be a day for the break. One more big battle awaits, and having come through 4 big days, the big teams are likely to look at rest and conserve energy for the Mont Ventoux 'showdown'. So don't expect too many fireworks, other than those provided by riders who will see tomorrow as their last shot at a stage win.

Preview of Mont Ventoux to come, and then obviously the race report on Saturday

Join us then!


Anonymous said...

This is a very nice blog and i'm enjoying reading both the analysis and the comments.
There has been some arguing that Contador was paced by motorcycles on both Verbier and at the TT today. I have a hard time seeing the difference it might have made on Verbier (other than having a wheel to follow) but in the TT pace and hence reduced headwind could really make a difference. This could explain why he was suddently secondbest on the down hill part of todays TT? futher more it has been said that he is not very strong at downhill? futhermore he is a very light rider? Could pace explain his very fast time and especially fast downhill? Can some analysis be done on pace?

Anonymous said...

Anon : Please look to Contador's past TT performances. This guy is a trailblazer. Cancellara needs to shut his mouth.

Anonymous said...

Hi Ross,

As always I enjoy reading your analyses and comments. Very informative and entertaining. I definitely agree with the comment above that the mass of the riders has played a major role in the splits you've described.

I believe Contador, the Schlecks and Wiggins are close to 60 kg while Cancellara is over 80 kg. On the Le Tour website the Cote de Bluffy climb is 234 m, and thus a 60 kg rider will require 255 W (+ power to over come drag and frictional losses) to climb this mountain in 9 minutes, the 80 kg rider 340 W! Going down the otherside Cancellara's extra weight would be a huge advantage, when cashing in on the potential energy he powered up the steep climb.

Interestingly on stage 17, the commentators were marvelling that Thor Hushovd was able to stay ahead of the bunch going down the big climbs towards his sprint points. But as with Cancellara, Thor is a big guy at 80+ kg so it's not surprising that he would descend faster than the peleton. What was surprising and rather remarkable given his added weight was his ability to attack and climb faster and get ahead in the first place.


Anonymous said...

The thoughts on pace by motorcycles relates more to the fact that Contador was some what faster than Klöden, wiggins and armstrong on the downhill and finishing sectors which might not have been expected. He obviously had a lot of photographers and cameras in front of him.
His TT before 2007 has not been that extraordinary.

Marco said...

It was also reported on french TV that the wind was stronger near the end of the day, making the first half of the TT easier and the second half harder for riders. That would explain some of the differences between Cancellara and GC contenders intermediate times.

Chris F said...

Wiggins said they hit a blockwind coming down off the climb which made all the GC contendors much slower over this section (as reflected in the times) than earlier starters like Millar and Cancellara. Think this also "equalises" the riders so time gaps narrow as they all struggle.

Anonymous said...

New to the site. Just wanted to say, you guys rock. Going to use this from time to time to make my math class more interesting.

Ron Wolf said...

um, didn't di Vinci demonstrate that heavier things don't drop any faster than light ones some 500 years ago? its about frontal surface area (drag) as well as weight and power. maybe you have data/results that show heavier riders descend faster, but that's not my experience. going up, on the other hand is a different matter. for one thing drag is less important at the lower speeds and power/weight far more important.

Ross Tucker and Jonathan Dugas said...

Hi All

sorry for the delay replying, just got round to these posts now. Thanks for the comments!

To Anonymous 1

I must say I didn't notice the pacing by the bike in the Time-trial. I guess it could make a difference. I don't know how one would quantify the size of the effect, because the bike has to have been at least 5 to 10m ahead of him. I don't know that the draft effect applies at that distance - if you think about it, if a cyclist loses the front wheel by even 3 m, the effect is negligible, so I can't see how a bike some 10m ahead (maybe 5) would provide that big a difference! Then again, I may be wrong - I don't know enough of the aerodynamics to say for sure...

Finally, I'm not sure what you mean by "pace" in this context? Sorry for misunderstanding...

To Anonymous 2:

I don't know about 'trailblazer' in the time-trials. Yes, he's a good cyclist, great even. Yes, he's absolutely untouchable in the mountains. But in 2007, he was limiting time losses in the time trials, now he is faster than the best time-triallists in the world. That's an exceptional improvement.

Having said that, I do think Cancellara's statements lacked a bit of class, but he was hurting and may have said some things in the moment. But I certainly don't know that even Contador's biggest fans would be predicting that he'd WIN the final, essentially flat time-trial of the Tour! Surely not?

To Frank:

Yes, the larger size does help when descending, because they produce greater absolute power outputs and since they're not carrying weight against gravity, this higher absolute power turns into much higher speed. Same effect as riding on a flat road, of course, and the reason why it's so unusual for the smaller climber to beat a time-trial specialist (who is larger) in the time-trial.

That 3km climb made a big difference - just from my splits, Contador was so much faster there, he really used it well. However, that first segment was amazing - 37 seconds faster than Cancellara! Even with the different strategies adopted by each, that's a pretty incredible performance by Contador!

To the next anonymous poster:

Quite correct - going back to 2007, he was certainly not a guy who you'd pick to win a time-trial! He has been on a steady curve upwards, but this was really an incredible ride, which I think everyone alluded to. There cannot be many people who watched that and said "predictable!". So you're right!

To Marco and Chris:

Thank you for that (and for all your other comments, great input!)

I didn't know much of the wind speed, it wasn't really reported on my TV feed. That fact makes Contador even more remarkable, if his performance happened in spite of more difficult conditions - that's amazing.

Normally, I'd ahve said that a headwind would blow the gaps wider, not narrow them. It's kind of like uphills separate strong from weak, whereas in this time-trial, the gaps seemed smaller. difficult to know if that was the wind or just the pacing of the guys catching up to them!

To 12th man:

Thanks a lot! That's very kind! And very novel! I hope your students enjoy it! Good luck!

To Ron:

Yes, you're right. The often held assumption is that heavier descends faster, but in the poster's defence, he didn't necessarily say it was a gravity-effect, which is where the mass is often implicitly assumed to act (it was Galileo, by the way, who showed this, supposedly by dropping objects off the leaning Tower of Pisa!).

So the effect is done to the power output produced, which is larger in bigger guys, and because that power does not have to overcome gravity (as on uphills), they go down faster. At least that has always been my understanding. I certainly don't have data on it - also, I think the commentators love to hype things up. I know that a lot of climbers are also described as great descenders, but that's probably a practice effect!


Anonymous said...

I thought heavier riders descend faster because air resistance is less of a problem for heavier objects. Gravity creates a bigger force on heavier objects (F = m*a remember?), but air resistance will still be about the same (given similar surface areas) so it troubles the heavier rider less. Or in other words: the resultant force created by gravity and air resistance together is relatively bigger for heavier riders.


Gravity working on light rider (60kg)
m*a = 60*10 = 600 Newton

Gravity working on heavy rider (80kg)
m*a = 80*10 = 800 Newton

Let's assume air resistance creates a backwards force of 600 Newton for both riders at a velocity of 60 km/u. Now it's easy to see that for the light rider the resultant forces cancel each other out completely so gravity does not help him anymore at that speed. But for the heavier rider gravity is still pulling for 200 Newton harder at him than the air resistance is pushing him back, so he will be able to go faster. :)

Ron Wolf said...

OMG, embarrassing.... yes, Galileo and the Tower of Pisa... and right, i inferred that the authors were talking gravity, my error (again ;-).

makes sense what you say about bigger guys putting more power to the pedals and therefore going faster downhill where gravity burdens them relatively less. makes sense, but i'm still not sure.

my observation is that large guys like to tout their superior TT and down hill ability, but that in practice the smaller guys actually do at least as well. Zabriskie for instance is not so heavy (67Kg according to Wikipedia) and fast as they come.

well, no biggie on this issue, more of a curiosity on my part if this had been corroborated.

btw, similarly (is it?) in running its a beginner's mistaken assumption that longer legs make for faster runners. in practice, not the case. there are many factors affecting efficiency that seem to actually penalize the larger runner. in the case of cycling, it seems to be easier for smaller guys to maintain higher cadence. so there is more to it than just pressure on the pedals.

Geoff Loveman said...

well, this is a little late but just in case anyone is reading this and getting confused about their homework on gravity....
Two objects with different masses will fall at the same speed under gravity in a vacuum. The gravitational force on the more massive body is greater but it takes a greater force to accelerate that more massive body. The less massive object experiences a smaller force from gravity but since it is less massive it doesn't need as much of a force to get it moving. Because of this, all objects accelerate at the same rate due to gravity but only in a vacuum. As soon as you consider drag, you will see that it slows down the lighter rider more than the heavier rider. So big boys can indeed descend faster for the same effort. Of course, on a technical descent, rider skill is probably much more important.