Core stability exercises such as planking, crunches and the Russian twist are widely incorporated into rehabilitation and sports performance programs to prevent injuries and improve sports performance. However, there are several misconception related to core stability training. In this post I will briefly discuss four of these misconceptions.

1.      Core stability training has a major impact on sports performance

In a recent meta-analysis, it was shown that classical core stability training has only a minor influence on sports performance in healthy young trained individuals (1). More specifically, the pooled effect size of core stability training on measures of athletic performance such as the time to complete a 5000-m run was 0.00 (no effect). Moreover, when the effect of core stability training on measures of athletic performance was compared to whole body strength training, the pooled effect size was -0.12, indicating that whole body strength training is slightly more effective than classical core stability training. Therefore, classical core stability exercises such as planks and crunches have only a minor impact on sports performance and can better be replaced by other exercises, especially exercises in which the emphasize is on intermuscular coordination rather than maximum force production of the core muscles.

2.      Core muscles can and should be selectively activated

In the field of physiotherapy, it is often thought that specific exercises can selectively activate some deep or local core muscles such as the multifidus or the transversus abdominis and it is assumed that these exercises are necessary to prevent muscle imbalances, selective atrophy or deficits in the activation. However, in a recent review, the authors found no evidence for exercises that could selectively target the deeper core or segmented stabilizing muscles and they found no evidence on selective atrophy or selective deficits in the activation in athletes (2). Moreover, several studies have actually shown that the core muscles are activated in a synergistic way during motor tasks, with the specific activation pattern depending on the exact motor task. Therefore, selective core muscle training of the deep and global muscles is –if possible- not justified and contrasts the synergistic functioning of both deep and global muscles during motor tasks. Moreover, such selective core muscle training may even be detrimental to the intermuscular coordination.

3.      Trunk rotation exercises are functional

‘Functional’ training usually includes some trunk rotation exercises such as the seated Russian twist. However, rotation exercises elicit high shear forces on the intervertebral disks when they are performed with a fixed hip and should therefore perhaps be avoided (2). Moreover, during many high-intensity motor tasks, the core muscles try to fix the trunk (i.e. limit rotational movements) in the upright position where the trunk muscles are the strongest and therefore it can be questioned whether the core muscles should be trained through rotational exercise. Training should be specific to the motor task and therefore it is perhaps better to improve the capability of the core muscles to fix the trunk, especially during high-intensity sports situations.

4.      Unstable surface core training is better than stable surface core training

Unstable surface training. Adapted from

It is assumed that core training on unstable surfaces puts higher demands on the coordination and requires higher muscle activity of the stabilizing core muscles. Unstable surface training is therefore supposed to be very effective for core stability training. However, to date no longitudinal study has been performed to validate this reasoning, which means that there is no evidence that shows the superiority of unstable surface core stability training. Similarly, there is also no evidence that unstable surface training leads to more muscle activity of the deep core muscles. Moreover, the load in unstable surface training has been reported to be approximately 40% lower compared to stable surface training, which makes it likely that the activity of most (core) muscles actually decreases (2). Finally, unstable surfaces are not used in most daily or sports situations and require a different motor control, which makes these exercises unspecific and perhaps a waste of time (see also my recent post on balance training using unstable surfaces for more information on this topic).

Conclusion and practical applications

Classical core stability training has only a minor effect on sports performance, probably because most exercises are not specific to the sports situation and because they do not put high demands on intermuscular coordination and force production. The slow and controlled movements in training do not replicate the fast and unexpected movements which occur during most sports situations. Furthermore, the core muscles are activated in a synergistic way during motor tasks and selective core muscle activation –if possible- should therefore be avoided. Finally, trunk rotation exercises and unstable surface core stability training are less ‘functional’ than commonly assumed and should perhaps also be avoided and replaced by exercises in which the core muscles are challenged by (unexpected) perturbations to fix the trunk as shown in the figure.

A step-up with an unstable load probably requires a high activation of the core muscles as they try to fix the trunk.


1.           Prieske O, Muehlbauer T, and Granacher U. The Role of Trunk Muscle Strength for Physical Fitness and Athletic Performance in Trained Individuals: A Systematic Review and Meta-Analysis. Sports Med 46: 401-419, 2016.

2.           Wirth K, Hartmann H, Mickel C, Szilvas E, Keiner M, and Sander A. Core Stability in Athletes: A Critical Analysis of Current Guidelines. Sports Med, 2016.

Share this post
4 misconceptions about core stability training
Tagged on:                 

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.