Lidl-Trek Head of Performance Josu Larrazabal explains how the team is embracing numbers and navigating a complex and changing world of data
Numbers and sport sparks endless discussion. There are thousands of ways to use and interpret data, and that murkiness creates separate debate between those who champion the value of numbers as a decision-making tool, and those who prefer to rely on impossible-to-measure traits using their eyes and intuition.
While all opinions may be (more or less) legitimate in an impassioned discussion, for those who have to make decisions, numbers cannot be ignored.
For coaches and sport scientists, numbers are like a photograph. They tell the facts as they are. They show us an athlete's performance through a simple language of recognized metrics - Josu Larrazabal, the architect of Lidl-Trek's performance structure.
Josu Larrazabal, team Head of Performance since 2017 (©SeanHardy)
The analytics era of cycling began with the advent of the power meter in the 1990s, followed by the emergence of software such as TrainingPeaks at the end of the decade. In the new millennium, the growth of technology was unstoppable, and numbers became vital to all the members of a performance team. “Like all professional sports, cycling has been hit by the avalanche of ‘big data’. Numbers and algorithms have become more and more central in our analyses and the need to dedicate resources has emerged,” continues Larrazabal.
To help deal with, and utilize, this avalanche of ‘big data’, Lidl-Trek expanded its performance team. The team’s first step was to recruit two expert data analysts, Xavi Tarralardona and Javi Angulo, the latter of whom is well known on social media as El Tio del Dato. The task of the data analysts is to extract data from across the whole gamut of rider performance, then arrange the numbers into a format that the coaches can easily interpret. Lidl-Trek also invested in two new coaches, Aritz Arberas and Xabi Zabalo, to help interpret this data and turn it into real-world performance gains.
The whole Performance Team of Lidl-Trek (©SeanHardy)
“This has been a significant upgrade for us,” reveals Larrazabal. “Data analysis has always been a fundamental element of our work, and now we have the opportunity to refine and extend the method. Before we were able to do day-by-day analysis in key races and moments of the season, but now we can do it for every race day of the men’s and women’s calendar,”
“Having two data analysts will allow us to convert the large amount of data we collect into different types of reports that will enhance the work of our coaches on their way to bring the riders to the top. For example, the race specific reports will support the team directors’ decisions at the races and the evolution reports could be used by the doctors or nutritionists in some specific situations. In short, they put the data at the service of every specialist in the most specific way possible.” says Larrazabal.
Lactate test is one of the many test carried during the training camp to collect data (©SeanHardy)
The system Larrazabal described is based on teamwork and connection between science and sport. The first step is the collection of data, carried out through sensors and high-tech devices. This is followed by the aggregation of data and the creation of algorithms that facilitate interpretation. Finally, the team studies the data to sift out answers.
So, numbers are an integral and indispensable part of the performance team’s daily work. But how much does data analysis weigh in the performance team’s decision-making process?
“The best scenario is where uncertainty is reduced to a minimum,” explains Larrazabal. “In the decision-making process, numbers are essential to offer certainties and facts, which help to reduce the margin of error in evaluating a performance or making a choice.
“The performance team and the riders have the same ambition: to hit the target. That’s the biggest challenge and it’s through an optimized process that we can face it,” affirms the Lidl-Trek Head of Performance. “Thanks to the numbers, we understand the effectiveness of training plans and the evolution of physical condition to perform at the best, at the right time. Data analysis offers the opportunity to monitor and find feedback, that for an athlete can become a mental confidence boost.”
Amanda Spratt stopping her Wahoo just after the finish line at Tour Down Under: race data are now saved (©CAuldPhoto)
Great rider performance doesn’t come by chance. It’s the result of a complex process that is repeated and becomes routine. Data helps riders determine what’s best for their training, providing them stability and security in their choices.
That said, data can’t tell us everything about a rider. Plenty of important factors can’t be measured, such as drive, leadership, race intelligence, and more. Lidl-Trek evaluates riders holistically; data is only a part.
“In performance, data counts – and will count more and more – but in the end the difference is made by the human side. It is functional to succeed and in difficult times of uncertainty could be even the driving factor, but in the key moments the difference between equal riders will be in their head,” Larrazabal said.
The human element is an impossible factor to decipher. Concepts such as talent, motivation, capacity to suffer, courage, competitive spirit, intuition and creativity cannot be represented by numbers. These are what make a rider capable of performances that break records, that go beyond expectations, beyond the wattages measured in tests. Relying only on numbers is like putting a limit on an athlete's potential
It’s clear that the weight of data analysis in the decision-making process has become increasingly relevant in the daily work of team performance. The same applies to another related aspect, namely the evaluation of a potential new recruit, especially a young one.
“Training and race performance numbers allow us to make an increasingly broad assessment,” Larrazabal said. “Using historical data collected throughout the Team’s history, we have created a database that supports us in comparing current data with some of the great riders who have raced on a Trek bike. Especially for young riders, this is an important added value, because through comparison we can better understand their potential.”
Group ride during our team camp in December (©RossBell)
As with the analysis of a performance, however, the number is not the determining factor in choosing an athlete. “It allows us to make a selection, to help our choice, but intuition is still a valuable skill,” Larrazabal said. “When we decided to sign Jonathan Milan, the performance numbers only served to confirm the opinion that we were dealing with a rider with a huge potential.
“I don’t think we will ever see a case history in cycling like the one told in the film Moneyball [the story of MLB team Oakland Athletics and its model of player selection based on advanced statistics], but the fact that there are riders in the peloton who come from the world of virtual platforms is significant,” Larrazabal said. “With big data and the data analysis process, our sport is evolving. Unlike other disciplines, such as arena sports like basketball, soccer or football, we maintain unique peculiarities. Riding on the road, in a peloton, for five or six hours, always has high uncertainty factors. Understanding how and where the science of numbers applied to cycling will evolve is a definitely fascinating subject.”
