Physics takes the plunge

How is the scientific study of top-level athletes' practice useful, especially here in swimming?

Science provides an understanding of the physics involved in a swimming event, and gives coaches and athletes alike the keys to improving their performance. As part of my thesis, I'm interested in the start phases, which are particularly important in a 50-meter freestyle, for example. The first 15 meters of the race alone account for 30% of the event's time, and not all swimmers use the same techniques. Some swimmers undulate more than others, with different strokes, and all these factors have an impact on the result at the finish. This means that for any given swimmer, at any given time, there is an optimal way to negotiate a start. We then need to study a set of parameters that the swimmer can develop with training. This is where science comes in.

How do you study these parameters?

With the goodwill of the athletes! We ask them to undergo three tests. The first is a start without pushing against the wall, during which the athlete accelerates as fast as possible up to maximum speed. This allows us to determine the propulsion and drag forces involved at the surface. In the second test, the swimmer is asked to undulate once he or she has reached a depth of one meter. It is then possible to determine the same forces involved, but underwater. Finally, in the third test, the athlete launches himself from a stud and has to let go as far as possible, enabling us to determine his ability to glide when passive. These experiments take place at the Institut national du sport, de l'expertise de la performance (Insep), where we have the in-pool video tracking system developed as part of the Sciences 2024 project. Thanks to a set of overhead and underwater cameras, it is possible to reconstitute a side view of the pool and record numerous parameters without disturbing the athlete. The videos captured are assembled and then processed by a network of neurons trained to identify sensitive data for swimmers. They are then analyzed and made available to sports teams.

What does this research show?

The first two tests revealed that very few athletes are capable of going as fast (or faster) underwater as they do above. By performing long underwater undulations, they end up going below their surface swimming speed. We then realized that some of our swimmers were spending too much time sinking in relation to their ability to propel themselves underwater. This observation undermines the idea that it's easier to move quickly beneath the surface, and therefore preferable to spend longer in the water. Coming right back up at the end of the first 15 meters is therefore not the best strategy. We were able to establish a physical model representing the swimmer's trajectory over this distance. It describes the start from the pad, a passive underwater swimming phase, an active underwater phase and then a surface swimming phase.

What prospects does your work open up?

At present, our model only guides swimmers towards a starting strategy. To refine it and be able to deliver "the right formula", we're going to compare it with the techniques used by Florent Manaudou, Maxime Grousset and Mélanie Hénique by carrying out a series of tests and measurements with them. Indeed, in view of their performances, the starting techniques of these elite swimmers are considered to be optimized. We'll be able to validate our work when the algorithm confirms that their methods are optimal. Our model will make physical sense, and it will be possible to propose it to athletes who have not yet perfected their starting strategy.

*LadHyX: a joint research unit CNRS, École Polytechnique, Institut Polytechnique de Paris, 91120 Palaiseau, France

** Thomas Brunel is also attached to the Saint-Venant hydraulics laboratory, a joint venture between EDF R&D and Ecole des Ponts Paris Tech.