Skilled biking has made big strides in turning into a extra trendy sport over the previous 15 years and the XDS Astana staff are a first-rate instance of how the easy step of embracing expertise can flip a last-placed staff into a serious participant, even with no celebrity rider like Tadej Pogačar, Remco Evenepoel or Mathieu van der Poel.
For the previous few years, XDS Astana have been one of many groups susceptible to relegation from the WorldTour, however they’ve adopted the highest groups in incorporating information science into their programmes, hiring information scientist Morgan Saussine to assist flip the staff’s fortunes round.
It is not a brand new idea – UAE Staff Emirates-XRG teamed up with the massive information analytics firm Presight and now have a bot known as ‘Ana’ to assist reply questions on easy methods to get the most effective out of their riders. Visma-Lease a Bike use AI to assist with diet and Purple Bull-Bora-Hansgrohe are speaking about incorporating all method of knowledge into one unified system, in keeping with a current report in Velo.
Even the lower-ranked groups are getting on board – Tudor Professional Biking have posted a job opening for a full-time Information Scientist – so incorporating information into professional biking is clearly necessary, and that is as a result of it really works. Working example, XDS Astana.
Astana had been on the decline ever since 2021, once they misplaced Premier Tech as a sponsor. They completed final among the many WorldTour groups in 2022, 2023 and 2024 and are behind ProTeams Israel-Premier Tech and Lotto, that means they need to beat two WorldTour groups to make the highest 18 within the three-year UCI Staff Rankings – a requirement to stay within the sport’s prime tier
At the beginning of the 2025 season, the staff wanted to attain about 5,000 factors greater than groups like Cofidis and Arkéa-B&B Accommodations over the season to make the highest 18. So Saussine used the out there information and helped suggested the staff on deciding on races the place their riders have the most effective likelihood at scoring factors.
Now, simply 18 weeks into the season, XDS Astana are third behind UAE Staff Emirates XRG and Lidl-Trek for 2025 thus far, and have succeeded in surpassing Arkéa-B&B Accommodations and Picnic-PostNL and are closing in on Cofidis within the three-year rankings.
The newest race content material, interviews, options, opinions and skilled shopping for guides, direct to your inbox!
The riders, coaches, sport administrators, mechanics and administration all contributed to XDS Astana’s success, however Saussine helped them not simply with their calendar and rosters, additionally with coaching and diet by creating instruments that may assist the staff make knowledgeable selections.
Saussine emphasises that his strategy is simply serving to human beings to make the most effective selections, and isn’t turning the staff into robots.
“The day when you’ll have an computerized AI voice that is gonna communicate in your staff automotive – I do not need to see it both. I just like the hybrid means of getting somebody to work with, as a result of we’re speaking about individuals who have years of expertise in biking, so I discovered so much from them,” Saussine advised Cyclingnews.
“In a WorldTour staff on the highest stage, when it is advisable to carry out, you could have totally different use instances on a weekly or typically each day foundation. So then you definately want actually somebody to concentrate on the information. It is fairly new in biking that it is advisable to do information engineering – gathering the information, do the extraction, remodeling, loading after which cleansing the information. Then you could have the information evaluation half, the place you may construct some dashboard charts to make the information extra readable for the top consumer. Then you could have additionally the information science half, the place you create fashions to make predictions.
“But it surely’s not PlayStation … [you have to make] it readable for the fellows who’ve the true experience – you will be the most effective coder on Earth, however when you do not perceive biking, it’s harder. I am a leisure bike owner – not at a excessive stage – however I used to be constructing instruments for myself, having a coach, doing testing… So I can communicate to a nutritionist, I can communicate to a coach, I can communicate to a DS – I’ve much less experience than them, however I can communicate their language.”
The variables you may management
The upward trajectory for XDS Astana additionally got here because of an infusion of money from the Chinese language bike maker XDS. The staff was capable of rent a slew of recent riders like Wout Poels, Sergio Higuita, and Diego Ulissi however the staff nonetheless lack a serious star like Pogačar, who has scored a 3rd of UAE Staff Emirates’ factors this 12 months, or Alpecin-Deceuninck with Van der Poel incomes practically half of the staff’s whole.
XDS Astana have as an alternative relied on their depth and data-informed selections. Their prime rider within the staff rankings, Simone Velasco, has earned simply over 10% of their factors. They’ve additionally balanced their targets nearly evenly between WorldTour races the place scoring factors is harder however extra profitable, Professional Sequence and .1-ranked races the place there are fewer factors however lighter fields.
Sports activities director Dario Cataldo noticed the outcomes of their work first-hand on the Presidential Tour of Turkey. The staff gained the general with Wout Poels, who took three phases, and had Harold López end second general.
“[Saussine] analyses the biking calendar [and predicts] the place we now have the best likelihood to attain factors with the riders we now have, bearing in mind their traits. In a sure interval of the 12 months, he research which race we must always participate in and with which kind of rider. It is bringing wonderful outcomes,” Cataldo just lately advised Cyclingnews correspondent Jean-François Quénet.
“It is a staff effort. It is not the information scientist who decides and the game administrators who observe. When there is a doubt, we offer him with data, we ask him to analyse the scenario, and he comes again with numbers that inform us what fits us. It is very helpful. In relation to his suggestions, we determine which riders we ship to which race in keeping with the likelihood of scoring factors.”
Because the begin of the 2025 season, XDS Astana has not solely worn out their deficit to 18th place within the three-year staff rankings, they’re on a path to climb even larger up the rankings in the event that they proceed on as they’ve within the spring. Might information analytics be the following ‘marginal acquire’ for biking?
Vasilis Anastopoulos, Astana’s Head of Efficiency, thinks so. “Morgan performed and nonetheless performs a big position in our choice making, concerning the roster of the staff, the number of the races. He defines the strengths and limiters of our guys,” he advised Cyclingnews.
“He is serving to us by selecting the most effective roster for every race, and in addition he is serving to us analyse the information after the races and point out if a rider wants a relaxation, if he is in an overload scenario, if the coaching we now have utilized to him is efficient or not, and that was a giant, large assist for us – an enormous asset to the staff over final 12 months.”
In fact, information evaluation cannot remedy every part. Christian Scaroni was the staff’s prime scorer till Strade Bianche, when he crashed closely, exhibiting that dangerous luck will be round any nook.
“Despite the fact that you possibly can choose the riders which have probably the most likelihood to attain with … if you do not have the legs on the day, when you crash on the moist highway, on the downhill – there are such a lot of variables that you don’t management. However that is what we love about biking too. We’re making an attempt to lower the quantity of unknown variables, however on the finish, you continue to have so much,” Anastopoulos mentioned.
Based on Anastopoulos, the following step for utilizing information is “to create some fashions predicting extra precisely issues like athletic efficiency and optimising the coaching program that we apply to the riders with the objective of the riders reaching their prime form on the moments that we wish.
“There’s nonetheless some work in progress, however the work that Morgan is doing and all the information that he is analysing is basically, actually helpful for choice making.”
