What does the data analytics partnership with Team DSM involve?
In essence, it is very simple: if you can generate better insights with data analysis, you can also take better decisions about the game plan. During the race, but also beforehand. In this data & analytics partnership, Team DSM and KPMG will be working together in order to generate even better insights in professional cycling.
Of course, cycling teams were using plans long before advanced data analysis was possible. But without input from data analysis, coaches and team leaders were forced to draw mainly on their experience and intuition when implementing those plans. With the right data, you can critically challenge those ‘gut feelings’ and – at least as importantly – have the right conversation about the best approach among professionals.
The use of data makes decisions at all levels more objective and better. From deciding which races a top cyclist should choose this year in view of their qualities, the course and other factors to choosing the best moment to attack during a stage.
Of course it is beyond dispute that data analysis cannot produce absolute truths. Data analysis improves the chances of success but does not guarantee it. You cannot be sure you will win a race with a particular strategy or develop the perfect one-year plan for a rider so that you know their form will peak at the right moment. But you can improve your chances of taking the best decisions and getting the best possible picture of the opportunities and risks during a race. So you can get a good sense of possible stage scenarios, the opportunities for the team or individual riders and the potential impact of strategic interventions. Always based on the plan.
Using data-generated insights makes decisions more explicit than previously. There are many areas where we can obtain such insights, and we systematically identify opportunities for doing so. Examples of how this benefits Team DSM include:
The ability to determine the optimum weight of a rider – not too heavy for mountain climbs but also not so light that it would compromise power. That also translates to precisely measured quantities (and composition) of food during and after a stage.
We can monitor how much force a rider applies to the pedals, allowing us to optimise energy management along with food and fluid intake during the race. This allows the team to get the maximum out of a rider without pushing too hard.
We can optimise tactics in a team time trial or mass sprint – including the configuration of the riders – by means of a combination of data from the wind tunnel, course information and physical data from riders.
We can optimally adjust the equipment to the rider – for example, the optimum position on the bike: not too hunched because that means a rider cannot deliver sufficient power, not too upright because that is detrimental in terms of aerodynamics.
In the future we will also be able to obtain instant information about the physical performance of riders and weigh that up against their rivals and the course. This will make it possible to take better decisions during the stage itself, or in any event to have the right conversation about what the best decision is
We can ‘map’ the data about riders’ qualities onto the course and the qualities of rivals and so make the best choices about participation in races (and the composition of the team).