a simple intermittent exercise analysis: gpexe vs traditional metrics (part 1)

Apr 4, 2019 | ACADEMY

What do we get from the accelerations number?

In this article, we will discuss the limits of acceleration events to describe the high-intensity phases in team sports and which gpexe metric can be used instead.

The exercise analyzed takes place over the entire length of the pitch and consists in a sequence of three accelerations/decelerations from the beginning to the end of the pitch; each “fast and loose” sequence is followed by a 30-s recovery phase and repeated for a total of six times (see image).

gpexe metrics vs traditional


The number of accelerations is often used as a measure of the “high intensity” volume that the athlete performs in training or in the game. Two parameters are required to define this number: an acceleration threshold and a minimum dwell time above the threshold. Most clubs use a threshold of 2.0 m⋅s⁻² for a period of at least 0.5 seconds.

A first consideration to make concerns the choice of threshold and time, that is completely arbitrary. In fact, there are no compelling reasons that justify precise values from a physiological point of view. In support of this consideration, just think of the fact that in the bibliography, there is no consensus on this topic. The numerous papers, that present acceleration, report very different thresholds and times. Needless to say, for the same exercise, these values are lower (both threshold and time), the greater the number of accelerations observed.

An even more critical aspect concerns the player’s ability to express high accelerations. Certainly, a high acceleration depends on the subject’s abilities to produce high powers and on the expressed commitment. A powerful subject, with good abilities to express strength, is more “accelerative” than a subject with lower ability to express force. Yet the comparison takes place using the same criterion (same threshold and same time)!

The other variable, in some ways even more hidden, concerns the starting speed that precedes an acceleration. It is very easy to produce a high acceleration from low speeds (very easy to make the tyres of a car slip from a starting position at traffic lights, by pressing the accelerator flat down!), but it is extraordinarily difficult to produce a high acceleration from high-speed situations (it is practically impossible to make the tires slip by giving gas when one is already travelling on the highway at 130 km⋅h⁻¹!).

In the field, these two situations are frequent and depend on numerous factors (the role, the style of play, the nature of the opponents, the terrain, …). Determining the number of accelerations without considering “how” these were made (what was my starting speed?) is a fairly useless exercise … and moreover, potentially misleading.

The exercise below is perfect to understand these concepts (the graph shows the time course of the speed):

gpexe metrics vs traditional

It would be reasonable to expect 3 accelerations at each repetition of the 6 bouts in the figure. Instead, the accelerations identified (flags on the X-axis) are about half, or 10 out of 18. Of course, the acceleration at the beginning of each bout is determined in all 6 repetitions: in fact, the athlete starts from a standstill and it is very easy to overcome the acceleration threshold. The same thing does not happen for the successive accelerations in which the subject starts at a speed of about 3 m⋅s⁻¹ (about 11 km⋅h⁻¹) and, obviously, can hardly exceed the acceleration threshold!

In this specific situation, if the necessary information is to identify the number of “high intensity” actions, the events of high metabolic power (power events) are much more suited to the purpose. Power events take speed and acceleration into account at the same time. The result is that, in the same exercise, the high-power events are exactly 18, just as expected.

Author: Cristian Osgnach

Related Contents
Related Contents