Changes of direction: myth or reality?

Introduction

Team sports can have very different demands, with some requiring frequent sprints and high-speed running [1, 2]. In contrast, others often involve lateral movements and changes of direction (CODs) performed at lower speeds [1]. These latter movements are particularly interesting to fitness coaches and trainers as they are highly demanding, especially from a neuro-muscular perspective.

CODs are most prevalent in soccer, while lateral movements are more frequent in basketball. Studies on soccer show that a player can perform up to 300 CODs in a match [1]. However, specialised literature lacks an unequivocal definition to recognise, quantify, and compare these movements in terms of trajectory and intensity.

Some studies [3] agree that CODs, defined as the body’s rotation in space and the speed of approach, are critical factors impacting the muscle load of the lower limbs. It is commonly believed that the angle of the COD (45° vs. 90° vs. 180°) can significantly affect the work done for the same distance covered. Regarding this aspect, Zamparo et al. [4] highlighted that at the same speed, different turning angles do not result in significant differences in energy expenditure compared to mere deceleration and acceleration in the same direction. Other studies [3, 5, 6] have focused on the speeds of COD execution based on the angle, measuring decelerations and accelerations entering and exiting the turn. As the angle becomes sharper, the time an athlete takes to complete the COD increases, leading to more intense decelerations and accelerations and thus suggesting a more significant muscular effort.

Given this, the question arises of whether a trainer must consider CODs during a match or training to assess an athlete’s training load more accurately and what criteria should be used to classify them.

By attempting to synthesise various approaches and disregarding the individual athlete’s skill in performing the movement and the different game situations, a COD can be reasonably considered a technical gesture consisting of three phases:

1. a braking phase approaching the point of direction change,
2. the actual change of direction (i.e., the rotation of the body in the new running direction),
3. a bursting phase exiting the point of direction change.

Investigation

With these premises, we organised a small investigation protocol to understand if some criteria can help to define COD actions and if there is any utility in assessing players’ workload in more detail by using this additional information. It’s often reported in the literature [7] that accelerations and decelerations greater than ±2.5 m⋅s⁻² are considered high-intensity actions. It is worth noting that the approach based on acceleration thresholds alone to determine high-intensity actions is less effective than the mechanical power-based approach [8, 9], mainly because the acceleration capacity decreases with increasing velocity and cannot provide satisfying results by using a fixed acceleration value.

Despite the abovementioned limits, we applied this criterion to detect high-intensity COD events by monitoring the entry and exit deceleration/acceleration. A subject (an amateur football player) was asked to perform a series of 10 shuttle runs with 180° CODs wearing a gpexe LT. Shuttle runs have progressed at increasing speeds up to the maximum possible over distances of 5+5 m and 10+10 m.

A team of 10 football coaches reviewed the videos and judged which of the ten shuttle runs represented high-intensity CODs based on their training experience. These evaluations were compared with the acceleration/deceleration measure above ±2.5 m⋅s⁻² from 5+5 and 10+10 m shuttles (Figure 1).

Figure 1 – upper panel: time course of speed during two CODs; the three phases mentioned above are indicated; lower panel: corresponding acceleration; the high-intensity zones are highlighted by the green areas above +2.5 m⋅s⁻² for acceleration and below -2.5 m⋅s⁻² for deceleration.

The experts’ evaluation did not always align with the results obtained from the acceleration threshold approach; thus, we expanded the investigation to include external work and power values calculated during CODs. In 2019, Zamparo et al. scrutinised the mechanical output during 180° CODs at different distances and speeds. Unsurprisingly, this study found that the work per unit of distance increases with speed and decreases in shorter distances. Moreover, mechanical power in most decelerations was about one and a half times the maximum acceleration power, confirming greater muscular capacity for eccentric activities.

To explore the characteristics of the entire deceleration/acceleration phase and more thoroughly distinguish CODs, we calculate positive and negative external power to obtain the external work performed (Figure 2). By doing so, we expected a more accurate indication of the muscular effort in this specific movement.

Figure 2: two of the ten shuttle runs performed by the player are shown, highlighting the speed (blue), the positive external power (green), and the negative external power (red) curves. To evaluate the intensity of the COD, it seemed reasonable to calculate the mechanical work done performing it, which is the time integral of the external power curves.

However, this approach poses the problem of identifying the time window straddling the rotation point to quantify the negative and the positive work. The longer this time, the greater the work performed, not ensuring a precise indicator of intensity and muscular effort of the analysed action. A time window of 1 second was established for the deceleration phase. This value was chosen based on Bloomfield et al. [10], showing that the average duration of deceleration events in a football match is 0.82 seconds, with 72.2% lasting less than one second and 95.5% less than two seconds. In the first instance, selecting 1 s as the braking phase seemed short enough to discriminate between deceleration events requiring high muscular effort and less demanding deceleration phases.

Besides the braking time, it was necessary to define the amount of work performed during the deceleration and acceleration phases to have a reference load for defining a high-intensity COD. The data so obtained from the tracker were compared with the expert opinions. The 10-metre shuttle revealed that from the 5th COD onwards, the external work exceeded the 6 J⋅kg⁻¹ threshold. The last five shuttles were precisely those judged by the experts as the high-intensity CODs. This amount of mechanical work aligns with the 5 J⋅kg⁻¹ cut-off proposed by Zamparo et al. [11] to define the most demanding CODs.

A shorter time of 0.5 seconds was considered for the subsequent acceleration phase. This allowed us to separate the quick-bursting phase following the trunk’s rotation (the actual change of direction), ignoring the following movements that could be subject to various game variables. The selected mechanical work reference for the bursting phase was equal to 3 J⋅kg⁻¹, maintaining a 2:1 ratio between the entry and exit phases for both work performed and the time required (Figure 3).

Figure 3: negative mechanical work during the braking phase (red area) and positive mechanical work during the bursting phase (green area) are highlighted according to the corresponding time windows (0.7 and 0.5 s for the braking and bursting phases, respectively).

Results

In summary, there was a correspondence between actions judged as high-intensity CODs from the video footage by the experts and those detected thanks to the mechanical work threshold of 6 J⋅kg⁻¹ in the 1s time window. Notably, deceleration times were always less than one second and generally lower than the values reported by Bloomfield et al. [10], averaging 0.82 seconds for more intense actions. This suggested a more appropriate deceleration window for significant intensity CODs equal to 0.7 seconds.

For the acceleration phase, the video judgment data also corresponded to the hypothesised target parameters for defining a high-intensity COD. Specifically, from the fourth shuttle run onwards, values exceeded 3 J⋅kg⁻¹ within 0.5 seconds after the change of direction. However, in the 5+5 m shuttles, there was no correspondence between the expert team’s judgment and the work performed. In this case, no shuttle reached the 6 J⋅kg⁻¹ target value for the braking phase, whereas almost all shuttles overcame 3 J⋅kg⁻¹ in the following bursting phase.

When acceleration and deceleration values were analysed to estimate the COD intensity, nearly no 5+5 m shuttles reached the ±2.5 m⋅s⁻² threshold. Again, similar to the work analysis approach, the 5-metre distance was too short for the subject to perform high-intensity deceleration and acceleration actions. The 10+10 m shuttles fared better, as the target intensities of ±2.5 m⋅s⁻² were achieved from the fifth shuttle run onwards, corresponding to the expert observers’ judgment.

Once the target values for detecting CODs were defined (0.7 s for the breaking phase, 0.5 s for the bursting phase, and 6 and 3 J⋅kg⁻¹ mechanical work, respectively), we applied the same criterion to detect the number of CODs and the respective angle performed by four elite players wearing the gpexe pro² tracker during an official match. On average, each player performed about fifteen high-intensity CODs with different angles and nearly 100 deceleration-acceleration phases with angles close to zero (i.e., occurring along the same running line).

Contrary to our expectations, the actual changes of direction, including trunk rotations, were few among the analysed players. Furthermore, considering that some studies indicate that the trunk angle during a COD does not significantly impact power and energy cost [4], it may be reasonable to consider CODs more as a technical aspect rather than a mechanical overload.

Finally, it is essential to note that the thresholds of 6 and 3 J⋅kg⁻¹ for defining a high-intensity COD are absolute values that do not consider individual capacities. Identifying thresholds relative to the athlete’s maximum power output would be more informative.

Conclusion

At the end of this simple experience, the question remains whether it is indispensable to record CODs during matches or training sessions, considering that:

• gpexe system already accounts for the most demanding accelerative and decelerative actions (i.e., bursts and brakes, respectively);
• the middle phase of the actual COD (the rotation of the body in the new running direction) does not seem to significantly impact the energy cost and mechanical power output [4];
• the thresholds set for defining high-intensity CODs are arbitrary and do not take into account individual athlete characteristics;
• the literature presents various criteria for defining CODs, focusing on different aspects, so it remains very complicated to have a standard way of defining a change of direction.

From the observations made so far, it can be concluded that in small-sided games, where the goal would seem to be increasing significantly the muscular workload, the amount of mechanical work achieved may not guarantee a significant power output and should be considered somewhat of an overrated exercise to provide a significant neuro-muscular stimulus.

Bibliography

1. Activity Demands During Multi-Directional Team Sports: A Systematic Review; J. B. Taylor, A. A. Wright, S. L. Dischiavi, M. A. Townsend, A. R. Marmon; Sports Med (2017) 47:2533–2551
2. The influence of soccer playing actions on the recovery kinetics after a soccer match; M. Nedelec, A. McCall, C. Carling, F. Legall, S. Berthoin, G. Dupont; Journal of Strenght and Conditioning Research, 2014, 28(6), 1517-1523
3. The Effect of Angle and Velocity on Change of Direction Biomechanics: An Angle‑Velocity Trade‑Off. T. Dos’Santos, C. Thomas, P. Comfort, P. A. Jones. Sports Medicine (2018) 48:2235–2253
4. Energetics of shuttle runs: the effects of distance and change of direction. P. Zamparo, I. Zadro, S. Lazzer, M. Beato, L. Sepulcri. Int J Sports Physiol Perform. 2014 Nov; 9(6):1033-9
5. Biomechanical analysis of two change of direction manoeuvres while running; P. Schot, J. Dart, M. Schuh; Journal of Orthopaedic & Sports Physical Therapy. 1995;22(6):254–8.
6. Change of direction speed in soccer: How much braking is enough?; K. Hader, D. Palazzi, M. Buchheit; Kinesiology 47(2015)1:67-74
7. High-intensity acceleration and deceleration demands in elite team sports competitive match play: a systematic review and meta-analysis of observational studies; D. J. Harper, C. Carling, J. Klely; Sport Medicine 2019
8. Mechanical and metabolic power in accelerated running-Part II: team sports. Osgnach C, di Prampero PE, Zamparo P, Morin JB, Pavei G. Eur J Appl Physiol. 2024 Feb;124(2):417-431.
9. Mechanical and Metabolic Power in Accelerated Running-PART I: the 100-m dash. di Prampero PE, Osgnach C, Morin JB, Zamparo P, Pavei G. Eur J Appl Physiol. 2023 Nov;123(11):2473-2481
10. Physical Demands of Different Positions in FA Premier League Soccer; J. Bloomfield, R. Polman, P. O’Donoghue; Journal of Sports Science and Medicine 2007a; 6(1):63-70
11. Mechanical work in shuttle running as a function of speed and distance: Implications  for power and efficiency. Zamparo P, Pavei G, Monte A, Nardello F, Otsu T, Numazu N, Fujii N, Minetti AE. Hum Mov Sci. 2019 Aug;66:487-496.

Author: Ivan Zadro