According to training software Inscyd, as used by WorldTour teams like Bora-Hansgrohe and Alpecin-Deceuninck, Jasper Philipsen generates up to 1,900 watts when sprinting to what the 25-year-old Belgium hopes is another stage victory. Similar figures have been quoted for former tornados André Greipel and Marcel Kittel, while Mark Cavendish’s maximum sprint power output is mooted at no more than 1,500 watts, the shortfall covered by not only strategic intelligence but also his low frontal profile. He is one savvy, slippery Manx. But what about the strain on the bonafide sprinters when the camera’s not dissecting their every movement? When they’re struggling up an Alpine or Pyrenean pass while the broomwagon looms? In other words, what are the physical demands on the grupetto to reach the finish line of a three-week Grand Tour? And how does it compare to the GC challengers
Hitting the sweetspot
Stage six of this year’s Tour de France will be remembered for Tadej Pogačar dropping Jonas Vingegaard en route to taking the day’s victory into Cauterets-Cambasque. The 144.9km parcours from Tarbes was one for the GC contenders and climbers, featuring two category one climbs plus the hors categorie brutality that is the 2,115m-high Col du Tourmalet. Its 17.1km rise averages 7.3% and witnessed Jumbo-Visma terrorising the competition. That was upfront. Outback, the likes of Cavendish and Fabio Jakobsen (now both out of the race) fought to tame the Tourmalet at a sufficient speed to make the time cut (usually around seven to 25% of the winning time) without sacrificing precious energy reserves for the following day’s sprint into Bordeaux (where Philipsen edged a gear-misfiring Cavendish). The entire grupetto made the cut by six minutes… which was misjudged.
“That was five minutes too fast,” the experienced Cavendish joked in Cauterets-Cambasque, a smile etched over his boyish face. The 38-year-old has started 23 Grand Tours, finishing 13. His best overall finish came at this year’s Giro d’Italia where he finished 120th. He knows what it takes to hit the sweetspot between making the cut, not digging too deep into the red and maintaining enough freshness in the limbs for peak sprint performance.
And so does Ineos Grenadiers data scientist Teun Van Erp, who analysed the performance characteristics of a world-class sprinter competing in the Tour de France in research published in the International Journal of Sports Physiology and Performance back in February 2021.
“We actually had two papers published at the time,” Van Erp corrects as we pin him down at the Science & Cycling Conference in Bilbao, three days before the Grand Départ. “One paper looked at the sprint characteristics of a top sprinter, specifically in the closing stages, while the other one essentially looked at how hard the mountain stages are, and the Tour de France as a whole, for a sprinter. The sprinter was Marcel Kittel and it threw up some interesting data.”
Power and positioning
For the purposes of the here and now, we’re more focused on Kittel’s trials and tribulations over 21 stages and upwards of 3,000km. Still, while we have Van Erp, it’d be remiss not to firstly delve into his finish-chute results?
“We analysed Kittel’s data from when I worked with him at Argos-Shimano [2013-2014, now Team DSM-Firmenich] and for a season at Quick-Step [2016-2017]. We analysed the positions from 10 minutes out to 15 seconds out and were following up on a paper by Paolo Menaspa who showed that ideally you need to be in the top-10 in the final 1km.” The paper is here.
Marcel Kittel's data provided insight into what it took for the sprinters to complete the Tour (Alex Broadway/SWPix)
“We showed that his power output was higher at Argos-Shimano up to 30secs out than at Quick-Step but he was in a better position. This better positioning wasn’t surprising because at Shimano, usually the whole team was supporting him but at Quick-Step, it might be three or four as Dan Martin would be going for GC. Argos were also often at the front, meaning Kittel wouldn’t enjoy the same aerodynamic ‘bubble’ as slightly further back. However, because the positioning wasn’t as optimum, Kittel needed to generate greater levels of power when 30secs and 15secs from the line to take the win.”
Before analysing the power and positional data, Van Erp hoped to observe tangible differences between victory and defeat. “That wasn’t really possible with Kittel,” he says. “He won 14 out of 21 sprints and the data for the majority of the ones he didn’t win wasn’t overly useful because he was caught up in a crash.”
Reverse of GC
We digress. The grupetto. Strain. Power outputs. What did Van Erp discover when studying Kittel’s power data from the 2013 and 2014 editions of the Tour de France? “That the guy who wins the flat stages will always suffer in the mountains!” he says. “Well, all aside from Wout van Aert, of course.”
“What’s interesting is that sprinters often do the reverse of GC riders on a mountainous stage, in that they ride harder up the first mountain because they want to hang on as long as possible,” Van Erp continues. “You want a large grupetto because if you’re dropped and have near enough the whole stage to ride alone, you’re f****d. They’ll be fatigued come the last climb but usually they can slow down by then because they know they’ve made the cut-off.”
This rapid start is not just reserved for the vertiginous mountains. On stage 10 of this year’s Tour – 167.2km from Vulcania to Issoire – ASO labelled its profile as ‘hilly’. It contained five category climbs albeit four were category three and one category two. Still, with temperatures well above 30°C at roadside, Lotto Dstny sprinter Caleb Ewan was dropped on the climb of the Col de la Moréno just 7km in. “Definitely better ways to spend a birthday than 165km dropped,” the Australian reflected in an Instagram post. “Thanks to the boys for getting me home in time.” Ewan crossed the line 34:09mins down on stage winner Pello Bilbao in the final group of six, a group that also included Soudal-Quick-Step’s Jakobsen.
“It’s why I feel a GrandTour is harder for the sprinters than a GC rider or climber,” says Van Erp. “For them, every mountain is on the limit, while they obviously have to perform on the sprint stages.”
For those accustomed to the power-based software TrainingPeaks, you’ll be well aware of the Training Stress Score (TSS). Broadly, this takes into account the duration and intensity of a workout to arrive at a single estimate of the overall training load and physiological stress created by that training session. By definition, says TrainingPeaks, one hour spent at Functional Threshold Power (FTP) is equal to 100 points.
“While the sprinter finishes in the bottom 5% overall, his TSS is around 750 AU (arbitrary unit) compared to a GC contender who might finish in the top 5%,” Van Erp explains, before addin: “Compared to a GC contender during a Grand Tour, a sprinter spends five to 10% less time in low-intensity power-output zones [zones one and two] and five to 10% more in the highest-intensity zones [zones four and five].”
The grupetto arrives on final climb of stage nine of the 2023 Tour de France (Zac Williams/SWPix)
A sprinter’s toil is further confounded by playing the support role for the GC rider if, Van Erp says, the stage profile is predominantly flat before finishing with a stage-defining climb, similar to the 2023 Tour’s stage 13 between Chatillon-sur-Chalaronne and Grand Colombier, which is relatively horizontal until the hors-category climb of the Colombier (17.4km at 7.1% to reach the mountain-top finish at 1,501m). “Sometimes if you’re going for GC and sprints, a sprinter would have to play his part in sheltering the GC rider,” Van Erp says. “It forges team spirit.”
Of course, the sprinter’s pain, like any rider, is impacted by the sadistic tendencies of the race organiser. When it came to Van Erp’s study, he calculated that Kittel burnt through 79,275kJ in 2013 and 92,106kJ in 2014. That equates to a 16% greater load when Leeds (UK) hosted the Grand Départ compared to Corsica.
The 2013 race came in at 3,403.5km long with three medium mountain stages and seven high mountain stages with the 2,001m Port de Paliheres mountain signifying the roof of the Tour. In 2014 the route was longer at 3,660.5km and featured five medium mountain stages and six high mountains with stage 14’s ascent of the Col d’Izoard the highest point of the race at 2,360m high. So, a similarly painful parcours. The major difference, however, saw the 2013 edition host a seemingly inconceivable three times trials, one of which was the team version, while the 2014 chapter hosted just the one. Like many riders in a team, aside from GC contenders and time trial specialists, the sprinter will see the time trial as a chance to conserve energy.
Slow-twitch v fast-twitch
It's data like this that makes training a sprinter such a delicate balancing act. “You can help a sprinter lose weight with more endurance work and they’ll survive the stage in a better condition but they’ll lose their sprint power.”
At a muscular level, Van Erp continues, too much endurance can train the powerful fast-twitch fibres to be more akin to stamina-packed slow-twitch fibres, so increasing fatigue resistance but impacting peak power output. Then again, I recall discussing this muscular-endurance tightrope with Caleb Ewan, just before he won two stages of the Covid-affected 2020 Tour de France.
He refrained from too much protein and gym work as he plied on muscle mass easily. “I don’t train that much for the actual sprint, either,” he elaborated. “We’re fast-twitch riders. That’s why we’re sprinters. That means your endurance muscles probably aren’t great, which is why we often suffer on the climbs. For us, sprinting’s around 200m of a race. I might have ticked off 160 to 250km before then or ridden several mountainous stages. It’s why the majority of our training focuses on endurance than sprint.”
Caleb Ewan in the mountains at the 2022 Tour de France (Zac Williams/SWPix)
Critics would suggest too much from Ewan’s three-year dry spell without a Tour stage win. Still, he can enjoy a crumb of comfort from carrying 13 fewer kilograms than Kittel in his pomp (69kg versus 82kg), which left Kittel with a power-to-weight ratio off 4.9 watts per kilogramme, which Van Erp suggested is the minimum required to finish the Tour de France.
Which naturally leaves us dreaming of Cav and what might have been. His 70kg, 1.75m profile expends far less energy ascending the mountains, helped further by a sleeker aerodynamic profile compared to the likes of Kittel (yes, at WorldTour rider speeds, aerodynamics has an impact uphill as well as down). Cav clocking number 35 down the Champs-Élysées with fresher legs than his larger sprinting contemporaries? Who knows, maybe 2024? Or maybe not. I mean, for how much longer would any member of the grupetto club sprint up an early mountain, hang on for dear life before readying themselves for the high-speed dangers of the following day’s sprint? The easy life of the GC rider must make them sick…!
Cover photo by Alex Whitehead/SWPix