Introduction

In recent years youth tennis has become increasingly competitive across all levels and ages. Recognition of the importance of physical preparation and development in youth tennis has also grown. Therefore, effective modalities which improves physical attributes pertinent to performance is crucial. The serve is regarded as strategically the most important stroke. As a result, improving service performance is a major goal of youth tennis programmes. A key aspect of service performance is serve velocity. The serve involves multiple body segments simultaneously producing force via complex coordinated muscular activations; termed the ‘kinetic chain’. Fett, Ulbricht and Ferrauti (2018) found that in elite youth tennis players, strength and power predictors explained 41-66% of the variance in serve velocity in boys and 19-45% in girls. Therefore, to increase serve velocity and enhance tennis performance, youth programmes must aim to attenuate strength and power throughout the kinetic chain.

Plyometric training is a well-established, appropriate and safe modality for enhancing power, movement velocity and the performance of explosive actions in youth athletes. Plyometric exercise causes muscles to stretch and shorten rapidly – a phenomena named the ‘stretch-shortening cycle’ (SCC). Plyometric training induces neural adaptations resulting in an improved ability to utilise the SCC and generate greater tension and subsequent force. Plyometric training can provide specific stimuli via manipulation of movement patterns, velocities, loads and metabolic demands. Chu (2003) for example proposed guidelines for tennis-specific plyometric training. Despite the theoretical potential for transference to the serve, this has not been fully elucidated in the literature. Therefore, the aim of this critical evaluation is to review studies on plyometric training interventions on serve velocity in youth tennis players.

Papers were sourced using SPORT Discus, PubMed, Google Scholar and relevant references from acquired studies. Search terms included ‘plyometric training’, ‘serve velocity’, ‘youth’ and ‘tennis’. Boolean modifiers ‘AND’ & ‘OR’ were used to narrow the search and to include alternative phrases such as ‘plyometrics’, ‘service velocity’ and ‘young’. Studies were required to include a plyometric training intervention lasting at least six weeks, a control group, a sample of young tennis players and outcome measures including serve velocity.

Review

All but one of the studies measured peak serve velocity. Fernandez-Fernandez and Ellenbecker (2013), Fernandez-Fernandez, De Villarreal, Sanz-Rivas and Moya (2016) and Pardos-Mainer, Ustero-Perez and Gonzalo-Skok (2017) all had players perform eight maximal serves and used the highest recorded velocity for subsequent analyses – to be recorded, serves had to land in the service box. In contrast, Behringer et al. (2013) measured mean serve velocity, recorded over twenty serves with no accuracy requirements. Behringer and colleagues argue this enhances the study’s practical relevance and ecological validity. On one hand, fatigue during matches reduces serve velocity. Thus, it could be suggested that improving mean velocity over numerous serves is of greater importance than enhancing the maximal velocity of fewer serves. However, this method essentially measures the effects of plyometric training on fatigue resistance during repeated submaximal serve performance, rather than maximal serve velocity. Moreover, the lack of accuracy requirements negates the potential ecological benefits of this method. If a player can serve with greater mean velocity over 20 serves but can’t land in the service box, performance simply is not improved. Additionally, Behringer and colleagues’ findings are undermined by other limitations. This study reported a significant increase in serve velocity compared to control post-intervention. However, mean velocity decreased over the study period in the control group (–5. 3%), whilst simultaneously increasing in the plyometric group (2. 9%). This suggests that the significant difference reported by the authors was caused by external, uncontrolled variables and not solely by the intervention. Moreover, the authors did not report a significant within-group improvement from pre to post-intervention in the plyometric group. Thus, the results from the Behringer et al. (2013) must be interpreted with severe caution.

Behringer et al. (2013) and all of the other studies in this review did not blind the interventions. Blinding is a crucial methodological tool in reducing biases. Not utilising it can result in the Avis and/or John Henry effect. This is where participants assigned to the control condition are either disappointed and perform at a lower rate (John Henry Effect), or are motivated and outperform the intervention group (Avis effect). Whilst it is difficult to effectively ‘blind’ a plyometric intervention, it is possible to reduce these effects via manipulation of study design. For example, using certain repeated-measures designs, a waiting list control or a staggered baseline. However, the studies in this review did not employ any of these tools. This reduces the ability to interpret these studies’ findings as strong evidence for the effects of plyometric training on serve velocity. Furthermore, the John Henry Effect may partly explain the decline in serve velocity observed in the control group in the Behringer et al. (2013) study.

The intervention programmes in the majority of studies were similar and deemed suitable and appropriate for adaptation. Interventions lasted six to eight weeks, training two to three times a week with sessions including upper and lower-body exercises with familiarisation and coaching throughout. Fernandez-Fernandez et al. (2016) and Pardos-Mainer et al. (2017) utilised similar programs, adhering to traditional plyometric programming principals. These two studies reported mixed but relatively positive results. Behringer et al. (2013) based exercises on NSCA recommendations, a needs analysis by Reid et al. (2003) and tennis-specific plyometric training recommendations from Chu (2003). However, these exercises were then compiled into one upper and two lower-body ‘circles’ consisting of three to four exercises. These were then performed in an interval training style – unlike traditional plyometric training. Fernandez-Fernandez and Ellenbecker (2013) utilised a range of exercises including plyometric, core and elastic tubing and medicine-ball exercises. This intervention could be considered as not a strictly plyometric program, rendering the intervention unsuitable. It could be suggested that the adaptations from these two somewhat inappropriate interventions would not be consistent with plyometric training. Both these studies reported significant improvements in the plyometric group compared to control and post-intervention. However, it is difficult to interpret these studies as strong evidence for the effects of specifically plyometric training on serve velocity in youth tennis players, due to these methodological shortcomings. Neither Fernandez-Fernandez and Ellenbecker (2013), Pardos-Mainer et al. (2017) or Behringer et al. (2013) controlled training volume between the intervention and control group. Participants in these plyometric groups performed the program on top of the standard in-season regimen. Therefore, these participants were exposed to greater training volume compared to control. The improvements in serve velocity reported by these studies could have been due to the greater training volume, rather than the intervention alone – making the interpretation of the results of these studies difficult. On the other hand, Fernandez-Fernandez et al. (2016) did control the training volume between groups. This was achieved by having participants complete plyometric sessions as a substitute for periods of tennis training. This constitutes a simple and effective way of equalising training volume between groups. However, many coaches would prefer significant training time focussed on technical aspects of the game – especially with young players, for whom skill development is of paramount importance (Reid et al. , 2007). Therefore, implementing plyometric training in this way may not be suitable for youth tennis programmes; reducing the applicability of this study’s findings.

The predicted age at peak height velocity (PHV) is crucial when attempting to recognise if developments in power are caused by training interventions or by natural improvements that occur during maturation and growth. The beginning of this natural increase in power usually arises approximately a year to a year and half prior to PHV. Biological age of maturity can be calculated to identify how many years before or after PHV participants are at the time of measurement. Differences in biological age of maturity between groups can be a confounding variable, explaining some of the differences in serve velocity between control and plyometric groups. Therefore, it is crucial for studies to assess the sample’s biological age of maturity in order to control for these confounding paediatric factors. Behringer et al. (2013) utilised a self-rated pubertal stage test to measure maturational status. This produced nominal data, whereby participants were assigned to one of five pubertal stages. This method allows an idea of differnces in pubertal stage between groups but is limited in scope for statistical analyses. Pardos-Mainer et al. (2017) only measured age, height, weight and BMI. In contrast, Fernandez-Fernandez and Ellenbecker (2013) and Fernandez-Fernandez et al. (2016) measured maturity age and reported no significant differences between the control and plyometric group. This suggests that the improvements in serve velocity in this study are not caused by differences in maturity. However, underlying mechanisms and adaptations are still hypothetical due to the absence of direct physiological measures such as EMG. Resultantly, it is difficult to confidently interpret the changes in serve velocity as a consequence of plyometric training alone (even if maturity is controlled).

Conclusion

The aim of this critical evaluation was to review the effectiveness of using plyometric training as an intervention to increase serve velocity in youth tennis players. The majority of studies found plyometric training to be efficacious in increasing serve velocity. Only two studies reported a greater improvement in serve velocity compared to standard in-season training alone. However, no studies reported plyometric training as being be less effective. This suggests a role for plyometric training in not only increasing serve velocity but also improving variance in training programmes; enhancing motivation and program adherence. With the exception of Fernandez-Fernandez et al. (2016), there is a lot to be desired in terms of methodological quality across all of the reviewed studies. Issues regarding measures, volume control, intervention programmes and paediatric factors hinders one’s ability to synthesise conclusions from these studies. Future research should initially aim to consolidate evidence for the effects of plyometric training on serve velocity in youth tennis players with research of higher quality.

Practical Applications

There is sufficient evidence to support plyometric training as an efficacious intervention to increase serve velocity. Therefore, coaches should include plyometric training in programmes with the aim of improving service performance; a key indicator of success. Plyometric training may cause neural adaptations which lead to augmentations in force production, movement velocity and explosive actions. Plyometric exercises can be manipulated to improve the specificity of movement patterns, loads, metabolic demands and angular velocities of exercises. This allows coaches to increase transference to particular sporting movements – such as the tennis serve. Recommendations must be synthesised with caution due to the methodological limitations of the reviewed studies. Further research is required to create standardised recommendations. This will simultaneously enhance the ability of practitioners to implement interventions and aid researchers in investigation. Nonetheless, certain recommendations can be made based on programmes utilised in reviewed studies with greater methodological quality and guidelines from other authors. Interventions should last at least six weeks, containing two to three sessions per week. Sessions should include upper and lower-body exercises. These exercises should be performed with two to three sets of ten to fifteen repetitions with fifteen seconds to two minutes rest between sets, depending on the exercise. Movements should be non-loaded unless performing medicine-ball exercises, whereby a mass of 2kg is sufficient. Athletes should perform exercises at high intensity with the intention to move rapidly. Certain considerations must be observed – athletes must undergo familiarisation sessions, with a slow increase in training volume and intensity. Athletes must always perform a warm-up prior to exercise and a cool-down following.