By the numbers: What it takes to win at the men’s Ardennes Classics
With the Ardennes now behind us, the 2018 Spring Classics season has come to a close. Before we turn our attention to the Giro d’Italia, however, it’s worth looking back at the Ardennes Classics in some detail. In the following article, former pro Stephen Gallagher from…
Heading out the door? Read this article on the new Outside+ app available now on iOS devices for members! Download the app.
With the Ardennes now behind us, the 2018 Spring Classics season has come to a close. Before we turn our attention to the Giro d’Italia, however, it’s worth looking back at the Ardennes Classics in some detail.
In the following article, former pro Stephen Gallagher from Dig Deep Coaching breaks down each of the men’s Ardennes Classics — the Amstel Gold Race, Fleche Wallonne and Liege-Bastogne-Liege — analysing riders’ power data to show just how hard these one-day WorldTour races are, what it takes to be competitive, and what it takes to win.
Note: This article discusses the power outputs and power-to-weight ratios of professional cyclists. For more information on these concepts, and to see where you stack up, click here. This article also deals with the concept of normalised power. To read more about this concept, click here.
Amstel Gold
It is rare to have access to the power file of the winner of a major Classic. But the winner of the 2018 Amstel Gold Race, Michael Valgren (Astana), posted his full ride data to Strava. It took Valvgren 6 hours and 40 minutes to cover a gruelling 263km and 3,000m at an average speed of 39.4km/h.
The start of the race was typically aggressive for a major classic and the main break of nine riders got its chance after about 14km of racing. It eventually gained an enormous gap of 16 minutes before the bunch started to reel it back in. Although Valgren was not attacking in the start phase, it was hard for all riders in the opening kilometres just to stay in position and maintain contact with the bunch.
In the first 15km of racing, the Dane averaged 271W (3.78 W/kg) with a normalised power of 320W (4.44 W/kg) – a significant effort, even for a WorldTour pro. Even more impressive were the numbers of Lawson Craddock (EF Education First-Drapac), who was very aggressive at the start and made it into the break. He put out nearly 100W more in that opening section, averaging 365W (5.11 W/kg).
After the break had established its solid lead, it started to hit the first of the 35 climbs, requiring repeated efforts at 5 to 6 W/kg for 2-4 minutes, even in the opening three hours of racing. Valgren’s average power for the first 130km and 3:34 of racing was 208W (2.89 W/kg), or a normalised power of 235W (3.26 W/kg). In the break Craddock covered the same distance more than 15 minutes quicker and put out 287W (4.01 W/kg) on average and over 300W normalised. This is a massive difference in work rate and energy expenditure between riding in the bunch compared to the breakaway.
The pace in the field really started to ramp up at the 100km-to-go mark. The chase from the peloton intensified and the efforts on the climbs became noticeably higher. Starting with the Vrakelberg at around 96km to the finish, Valgren produced 520W (7.22 W/kg) for the 1:23 needed to cover the short and steep climb. Most climbs in this period were done at around 6-7 W/kg, most notably the second ascent of the Cauberg after 178km at 481W and the first ascent of the Bemelerberg after 194km at 469W, respectively.
This shows how the big teams increase the tempo over the course of a long classic.
The final of the race usually begins with the Kruisberg after 221km (44km to go) and features various tough climbs in quick sucession, among them the Eyserbosweg after 225km, the Keutenberg, and the third ascent of the Cauberg. To stay in contention, Valgren had to produce the following impressive efforts:
On the Keutenberg, with its steep section of over 20%, the riders in the main group attacked each other and chased hard to catch the breakaway. Out front Craddock lost 32 seconds on the 1.1km-long climb alone, climbing it in 3:12 at 436W (6.1 W/kg). In the meantime Valgren put out 508W (7.06 W/kg) to climb it with Valverde and co in just 2:43.
The pressure continued with big numbers on the Cauberg after 244km as the now significantly thinned-out group of favourites went over it for the last time. Here, Valgren did 595W (8.26 W/kg) for the 1:26 it took him to cover the steepest sector of the famous hill. This was over 100W more than on the previous two times up the climb and it brought a very select group — including Valgren, Peter Sagan, Alejandro Valverde and Roman Kreuziger — up to the remaining escapees, Craddock among them. The American had still produced 470W (6.57 W/kg) on the Cauberg after more than six hours at the front of the race.
Then followed the Bemelerberg with less than 10km to go, where Valverde put in a last desperate attack to gap his rivals. Valgren was able to follow with 536W (7.44 W/kg) for 1:35 on the main 800m of the climb. Even more impressively, Craddock was able to minimise his losses to the much fresher riders with a 1:40-long effort at 481W (6.73 W/kg). His ability to go so far over his threshold after nearly 250km of hard work would earn him ninth place.
The big play for the final victory came in the final 5km as Valgren made a number of attacks trying to forge clear. His two main attacks in this final section were as follows:
The successful second acceleration was a big punch and this 47-second effort was enough to create a significant gap between him and the others, with only Kreuziger able to follow. The Dane reached an average speed of 47km/h during this attack, on a 2% gradient!
From here to the finish it was all about maintaining the small advantage and making sure he had enough left for the sprint. The pace was high in the last 700m, which he rode at 521W (7.23 W/kg). Valgren and Kreuziger didn’t have time to play tactical games because they had to worry about Bahrain-Merida’s Enrico Gasparotto, who had started to chase with 2km to go.
Valgren positioned himself perfectly for the sprint, sitting on Kreuziger’s wheel with 200m to go and made his move with 150m to go. He had enough power left in the legs to take a hard-fought victory with a final sprint, producing an average power of 925W (12.85 W/kg) for nine seconds, maxing out at 1,050W. A huge last effort after 6.5 hours of racing.
La Flèche Wallonne
This mid-week classic separates Amstel and Liege-Bastogne-Liege and is always a great indicator to who will contest Liege on the Sunday. It is also a prestigious and extremely hard-fought race in its own right. Traditionally, the race has always been decided on the mythical hilltop finish of the Mur de Huy.
To analyse, we have the data from one of the main animators of the race, Jack Haig (Mitchelton-Scott), along with pre-race favourite and five-time winner Alejandro Valverde (Movistar).
Haig had a big day out and was extremely aggressive in the final 40km. He went on the attack with superstar Vincenzo Nibali (Bahrain-Merida) and they got a small group clear of the peloton, putting Valverde’s Movistar team under pressure and isolating the big favourite.
Haig had tried to get into a move on the first of the three ascents of the Mur de Huy with about 60km to go. He put out 478W (7.03 W/kg) for 3:19 as a large group tried to gain an advantage. He was soon brought back by Movistar’s domestiques. Some of the instigators kicked again on the next climb, the Côte de l’Ereffe, and this time Haig managed to ride away with a smaller group around Nibali:
Note the incredible speed by the Australian on a 7% gradient: nearly 28km/h! The group opened up a 20-second gap before the next climb, the Côte de Cherave, and suddenly, Valverde had just one helper left. Haig drove it hard over the top with Nibali and QuickStep’s Maximilian Schachmann and the gap grew to 30 seconds.
Haig put out 455W (6.69 W/kg) for just under four minutes as he sustained a hard tempo up the hill. They were even able to gain another six seconds on the second ascent of the Mur de Huy 6km later and with 25km to go. Haig climbed it in 3:56 at 6.63 W/kg compared to Valverde’s 4:02 at 6.54 W/kg.
A main conclusion for us mere mortals: Haig produced around 6.5 to 7 W/kg consistently every time the road sloped uphill in the last 50km, for efforts between one to four minutes in duration. This is where the best WorldTour professionals differ from even the strongest domestic pros. They are able to repeat these super-hard efforts several times without slowing down.
The gap grew to 40 seconds and it was only on the penultimate climb, the second time over the Côte de Cherave with 7km to go, that Valverde got some help from other teams and the advantage started to come down again. Still, the Spaniard had to invest a lot of energy, helping to drive the pace:
In the last 3.5km, Haig attacked the leading group to go clear with Schachmann. His initial jump saw him hit 681W, to get the gap and distance Nibali, among others. For the next 26 seconds he averaged 60.3km/h, which required him to ride at 413W (6.07 W/kg) on a slight downhill towards Huy. Unfortunately, Haig’s days were numbered and he was picked up by the charging bunch on the slopes of the Mur. A great showing by the Australian regardless, as he went on to finish 34th.
It came down to a battle of the best puncheurs on the final climb. The numbers of Valverde are a testament to how strong these riders are. On the final 1.2km, he averaged an eye-watering 513W (8.14 W/kg) for the 3:14 it took him to cover the full Mur de Huy. To battle it out for the win in the last 200m, his average power amounted to 626W (9.94 W/kg) for 55 seconds with the last 18 seconds of sprinting at 717W (11.38 W/kg).
But it wasn’t enough to collect another victory at the Flèche Wallone as Julian Alaphillipe (QuickStep Floors) had a bit more in his legs. The Frenchman climbed the Mur four seconds faster and we estimate his effort at about 8.3 W/kg for 3:10.
Liege-Bastogne-Liege
The last of the Spring Classics, “La Doyenne” (“The Old Lady”) is a particularly hard test. Many riders have had a full and stressful spring already and at Liege they clash with climbers and Grand Tour specialists testing their fitness before the Giro d’Italia. With its longer climbs of up to 4km in length, Liege is generally suited not just to the puncheurs, but also to climbers who can sustain hard efforts on steep gradients for more than five minutes.
A nine-man break formed on the Côte d’Embourg (4.8km at 4%), which welcomed the riders in the opening 5km. Anthony Perez (Cofidis) made it into the break, having featured in it during last year’s edition as well. He produced nearly all his peak powers in the first 30 minutes of the race.
The decisive acceleration over the Côte d’Embourg saw Perez put out 460W (6.57 W/kg) for 2:19. The final big jump into the move was his peak 10-second power of the day: 934W (13.34 W/kg) and a max power of 1,263W. He also produced his highest eight-minute power here — 426W (6.07 W/kg) for the full attack as he averaged 32.5km/h on the 4% gradient. A hard start to a long day.
The peloton did not let the breakaway go easily. Jack Haig, a revelation of this year’s Ardennes Classics, had to put out 390W (5.74 W/kg) on the Côte d’Embourg and crested the climb just 10 seconds behind the escapees before the field eased their chase.
The pace and effort did not drop for the nine escapees as they fought to extend their advantage over the peloton. In the first hour, Perez and his companions pushed the advantage out to a maximum of six minutes and the French rider averaged 292W (4.17 W/kg) and a normalised power of 321W (4.59 W/kg) in this time. This was a monstrous effort, considering the 5.5 hours of racing still to come.
Here are Perez’s overall stats for the day, compared with those of eventual second-place finisher Michael Woods (EF Education First-Drapac) and 14th-placed Haig:
After having averaged 349W (5.54 W/kg) for 8:41 on the opening 5km — including the Côte d’Embourg — Woods tried to save as much energy as he could for the end of the race. In the subsequent 50km that followed, Woods averaged 185W (2.94 W/kg) and normalised 225W (3.57 W/kg). This is nearly 1 W/kg less than Anthony Perez, showing how much harder it is being in the break out front.
As the race went on, the average power on the climbs started to rise, most of the work being done to maintain position within the peloton as the break was chased back.
The critical phase of the race started at the Côte de la Redoute, which the riders crested with 36km to go. While no one dared attack this far out, the high pace meant the race was over for the breakaway, which lost 40 seconds on the short climb. Perez produced 380W (5.43 W/kg) for 5:35 compared to Robert Gesink’s 466W (6.42 W/kg) for 4:55.
When the field hit the Côte de La Roche-aux-Faucons 15km later, the fireworks started. As soon as the bunch made the right-hand turn into the foot of the climb , significant attacks started. Woods rode the main 1.15km (10.5%) in 3:41 at 462W (7.33 W/kg), a very similar effort to that produced by Haig, who climbed it in 3:40 at 492W (7.32 W/kg). Both remained in the front group as eventual race winner Bob Jungels (QuickStep) Floors attacked and rode away.
On this climb and the drag which followed it, Woods produced his peak powers from 3:30 to 10 minutes, the latter being 358W (5.7 W/kg) with 240 tough kilometres already in his legs. He tried to catch Jungels on the uphill drag after the Roche-aux-Faucons, accelerating again at 446W (7.08 W/kg) for two minutes. This was the last time anyone saw Jungels, who went on to ride the last 20km solo to claim what is probably his biggest career victory.

The final 6.5km of Liege are super-hard, with the Côte de Saint-Nicolas (1.1km at 9.2%) followed by the final climb to Ans. When the group of chasers including Valverde, Woods and Haig started Saint-Nicolas, they were nearly 40 seconds behind Jungels, but an attack by Jelle Vanendert (Lotto Soudal) that Woods was able to follow brought the advantage down to just 25 seconds.
Woods achieved his three-minute peak power here, doing the 1.1km climb in 2:54 at 23.1km/h, his average power 480W (7.62 W/kg). In the meantime Haig had to let the group go, losing 20 seconds but still putting out 451W (6.63 W/kg). Amazingly, however, he was able to crawl his way back on the descent towards the final climb, making the most of a short lull in the chase group.
In the final 3km, the chance of victory had disappeared because Jungels was still more than 30 seconds up the road. Haig went straight to the front of the group for his teammate Roman Kreuziger, setting the pace on the first part of the Côte d’Ans with 479W (7.04 W/kg) for 1:25. Romain Bardet (Ag2r La Mondiale) had attacked with just under 3km to go, taking Woods with him. Woods surged at 741W (11.76 W/kg) for 10 seconds and they brought Vanendert back in the final kilometre.
In the final 1.3km, the Canadian Woods averaged 436W (6.92 W/kg) for 2:57, distancing Bardet with one last acceleration: 536W (8.51 W/kg) in the last 400m!
It proved to be a breakout ride for the Canadian climber, finishing second in one of the sport’s five Monuments in just his third season as a WorldTour pro. His power numbers, and those of the riders around him, showed just what it takes to be competitive at the Ardennes Classics: ridiculously high power-to-weight ratios, the ability to put in hard efforts over and over again, and quite often, a strong sprint after more than six hours of racing.
[ct_highlight_box_start]
About Dig Deep Coaching
Sign up to the Dig Deep Coaching Newsletter to receive a free PDF for masters riders. Keep getting faster as you get older with our structured workouts and techniques to improve.
Dig Deep Coaching is a global coaching company that works with athletes of all levels across the following disciplines: road, track, cyclocross, MTB and triathlon. Whether you are taking part in your first ever gran fondo or aiming to compete in the professional peloton, Dig Deep Coaching can help you out. Get in touch via email or follow Dig Deep Coaching at Facebook and Twitter.
[ct_highlight_box_end]