The ability to sprint is essential in the majority of team sports, including rugby, football, and basketball. But this ability is not developed in stable and controlled conditions: players are constantly subjected to variable demands, whether receiving, carrying, passing, hitting, or throwing the ball. Traditionally, athletes work on this aspect in the gym using “weights” in purely vertical conditions, but that has changed in recent years. Now, strength work adds challenges that contribute towards improving adaptability and with that, performance. However, to study the resulting variability, conventional linear measurements – such as acceleration – seem to fall short, especially since the data they can provide regarding these changes are quite limited.

Within the study of movement, a practice known as entropy analysis has gained popularity in recent years. This is a non-linear type of measurement that deals specifically with the variability or the “disorder” of a time series. Paradoxically, it has never been used to study strength training in team sports.

A team of researchers, including Jairo Vázquez, physical trainer for F.C. Barcelona, have developed a first-of-its-kind study that shows the ability of entropy measurements to capture variability in these types of exercises. The study is linked to a doctoral thesis that was completed at INEFC Barcelona and published in the Journal of Science and Medicine in Sport in collaboration with researchers from CIDESD from Trás-os-Montes e Alto Douro (Portugal) University.

A Pioneering Study

“It is increasingly clear that strength training should consist not only of exercises such as squats with a bar and some weights,” maintains Vázquez. “We have to add challenges or conditioning factors that allow for an adaptation to variability. This is because we don’t learn by constantly repeating the same solution to a movement problem, but by constantly solving a new movement problem. And a very simple challenge to introduce is the ball.”

In the study, twelve professional rugby players completed a strength exercise using a rotational inertia device, one combining a type of concentric and eccentric task, which has already demonstrated great value. Distributed over four different sessions during one week, the exercise consisted of a forward and backward movement on the horizontal plane to which the presence of a ball could be added. For that, the player had to catch a pass from another player to his right and then throw it towards the receiver situated to his left. “In football or in rugby, the majority of the movements are horizontal – movements that are not produced during typical squats. With this exercise, we mix several positive points: on the one hand, a type of movement that is more common in a rotational inertia device, one which we already know obtains positive results. On the other, the presence of the ball produces a disturbance that stimulates adaptability to the environment.” Through the use of an accelerometer, “we wanted to know the influence this disturbance has and what changes it introduces.”

Three types of measurements were taken. Some, the more traditional – mean acceleration and peak (or maximal) acceleration – are based on “more is better,” as Vázquez explains. Comparing the ball-assisted exercise with the no-ball exercise, there were no differences noted in peak acceleration, only in the average acceleration and the forward movement, which was greater when the ball was introduced.

The other, newer, measurements consisted of the entropy analysis, which measure “signal variability,” Vázquez states, since “each signal has a structure, and the acceleration does not have to be regular.” In this case, there were differences – or clear tendencies towards a difference – between the exercise with and without a ball, both within the forward and backward movement and overall. In addition, when using a type of analysis that addresses temporal scales (multiscale), variability appeared in the widest windows, “those that can be linked to movement,” specifies Vázquez.

According to the theory of motor learning, there has to be a specific level of variability – neither very large nor very small – for optimal performance. “It’s true that the results of the study were the ones we expected,” states the trainer, “but no one had ever demonstrated it before.” Additionally, it confirms that traditional measures are insufficient to capture this training variable.

A Benefit and One Step Forward

The adaptation to variability not only improve performance, but it also decreases the risk of injuries. “This is similar to the process by which a hand strikes a table,” explains Vázquez. “If we repeatedly strike our hand in the same place, the damage is big and localized. But if there is a slight variation in the contact surface, the damage is spread out and the risk is lower.”

Adaptation assumes a greater level of coordination, but this will reduce variability over time, and the training becomes less effective. “This research is the first big step toward continuing to study along this line and understanding its complete applicability to training. Now that we know we can measure variability, we have a tool to modify and adjust the exercises in order to place them in the window where we can train for adaptation,” he ensures. “The decrease in entropy would be the modern and equivalent signal that we have to add more weight on the machine.”

Only, in this case, weight means disorder.