Predicting repeat power ability through common field assessments: is repeat power ability a unique physical quality?

Introduction

Repeat power ability (RPA) can be described as the ability to repeatedly produce maximal or near maximal efforts against external resistance training loads (Natera, Cardinale & Keogh, 2020). The ability to maintain high levels of maximal power output is suggested to be important for performing repeated high intensity efforts (RHIE) in many sports (Apanukul, Suwannathada & Intiraporn, 2015; Gabbett & Wheeler, 2015; Gonzalo-Skok et al., 2016; Goods et al., 2022; Natera, Cardinale & Keogh, 2020; Schuster et al., 2018; Vachon et al., 2021). As such, there may be a concomitant need to effectively monitor training induced changes in RPA (Apanukul, Suwannathada & Intiraporn, 2015; Baker & Newton, 2007; Natera, Cardinale & Keogh, 2020; Schuster et al., 2018). To better understand and define RPA, and to determine appropriate training modalities that may be used to enhance RPA, a thorough analysis of associated physical qualities needs to be undertaken. The associations between neuromuscular qualities like maximal strength, maximal power output and speed are important to understand so training to enhance RPA can be emphasised and monitored accordingly (Baker & Newton, 2007; Mosey, 2011; Natera, Cardinale & Keogh, 2020). Similarly, the association between cardiorespiratory performance like aerobic capacity and repeat speed ability are also important to clearly understand RPA and the underpinning cardiorespiratory qualities that may determine RPA performance (Fry et al., 2014; Natera, Cardinale & Keogh, 2020; Sands et al., 2004).

RHIEs is a cluster of three or more high intensity sporting actions, like accelerating, cutting, tackling, and jumping with a maximum of 21 s rest between each high intensity effort (Spencer et al., 2004). RHIEs have been found to occur around critical periods of play and often lead to successful outcomes in team sports (Black & Gabbett, 2015; Gabbett & Gahan, 2016; Hulin et al., 2015). The ability to perform RHIEs is suggested to reflect the ability to maintain relatively high levels of maximal power output under some level of fatigue (Natera, Cardinale & Keogh, 2020). RPA assessments may provide useful information to determine RHIE performance. Despite the potential importance of RPA, there is little known about the underpinning physical qualities that are associated with or determine RPA.

Although direct investigations to determine RPA have not been conducted, several studies have looked at the relationships between anaerobic performance, as measured via a 30 s Wingate assessment, and various forms of repeated jump and repeated speed squat performance (Bosco, Luhtanen & Komi, 1983; Sands et al., 2004; Fry et al., 2014). Strong relationships have been reported between these RPA assessments, consisting of either continuous compliant rebound jumps or speed squats, and the 30 s Wingate assessment conducted on cycle ergometry (Bosco, Luhtanen & Komi, 1983; Sands et al., 2004; Fry et al., 2014). Despite the mechanical differences between these assessments, the strong relationship identified is likely due to both assessments requiring maximal intensity effort of the lower limbs over a similar duration of approximately 30 s.

Although several RPA assessments have been described in the literature, an assessment utilising 20 loaded counter movement jumps (LCMJ20) has recently been reported as a reliable RPA assessment that may better replicate sporting movements and can be easily administered in the field (Natera, Cardinale & Keogh, 2020; Natera et al., 2023). The LCMJ20 may be considered the most effective and reliable assessment of RPA, as the researchers examined a range of different jump types, measurement indices, calculations and methods of describing power decline in their study. The most reliable method to quantify RPA is to calculate peak power output for each jump of the LCMJ20 and then to calculate a percent decrement score (percent decrement = 100 × [total jump power/ideal jump power]−100) with the first and last jumps removed from the calculation (Natera et al., 2023).

With the current lack of research and understanding of the physical qualities that contribute to RPA, there is a need for further investigations. A greater understanding of the potential contributing physical qualities, may help practitioners design training programs to enhance RPA more specifically. On the other hand, RPA might lack any associations with common assessed physical qualities and this may suggest that RPA is somewhat unique and may require specific training in order to develop it.

The purpose of this research is to identify the underlying physical qualities that determine RPA. This study seeks to provide the practitioner and coach with a better understanding of RPA and the potential to target training towards the development of physical qualities that can best enhance RPA. We hypothesise that RPA will have substantial unexplained variance and that strong relationships will exist between RPA and RHIE performance.

Materials & Methods

Study design

A cross-sectional study design was used to investigate a range of field test assessments commonly used in team sports. This research was designed to align with the assessments and timing of assessments commonly used by practitioners working in team sport. Furthermore, the demands of the preseason training plan for the field hockey players participating in this study was accounted for. All participants were familiar with the assessments used in this study as they were routine assessments used by this particular cohort.

Following a further session of familiarization with the LCMJ20, participants were required to attend the strength and conditioning facility on four different occasions over a two-week period, two sessions in week one and two sessions in week two, with all sessions performed at the same time of day (Fig. 1). In session one, descriptive data was collected and countermovement jump (CMJ), isometric mid-thigh pull (IMTP) and half squats (HS) strength were assessed respectively with 20 min of rest between each assessment. During session two, which was conducted 72 h after session one, the LCMJ20 RPA assessment was performed. In the following week, session three included a 40 meter sprint (40 mST) and a repeated sprint assessment with a change of direction (RSA₁₈₀). Between each of these assessments, 30 min of recovery was provided. The Yoyo intermittent recovery test 2 (IRT2) was then conducted 72 h later in session four.

All of the running based assessments used in this study (40 mST, RSA₁₈₀ and IRT2) matched the protocols and requirements of the National Sporting Organisation, Hockey Australia and fulfilled part of their biannual player assessment batteries. As such all running assessments in this study were conducted on the same indoor tartan surface and all participants were familiar with the assessments used in this study.

Technical hockey training during the two week assessment period was modified to be low intensity, skill based training to facilitate recovery between sessions and to optimise both neuromuscular and cardiorespiratory performance for the assessments. A standardised warm up was conducted prior to the first assessment on each testing day. The warm up consisted of 5 min of cycling at a rate of perceived exertion of 3–4, followed by 2 min of dynamic flexibility exercises for the lower limb and trunk.

Results

Descriptive statistics for all variables used in correlation and regression analysis are presented in Table 1. Both HS_rel (Fig. 2) and RSA₁₈₀ (Fig. 3) had very strong, significant relationship with LCMJ20 (r = −0.728, p = 0.017 and r = 0.736, p = 0.015 respectively). All other variables exhibited trivial to weak non-significant relationships with LCMJ20 (ranging from r = −0.293 to 0.118) (Table 2).

Percent decrement in speed for RSA₁₈₀ was the only identified covariate in the predictor model of LCMJ20 with stepwise multiple regression analysis predicting 48.4% of the variance in LCMJ20, Adjusted R² = 0.484, F(1,8) = 9.43, p = .015.

Discussion

The aims of this study were to identify the underlying physical qualities that determine RPA. With a better understanding of RPA, practitioners wishing to develop RPA can better develop their exercise prescription to develop it more effectively.

The results of this study demonstrate that approximately half of the variance of RPA can be explained by IRT2, RSA₁₈₀, 40mST, IMTP_rel, CMJ and HS_rel. More specifically, the RPA variance explained is largely determined by RSA₁₈₀ percent decrement. The prediction of RPA only slightly improved with the addition of HS_rel. Overall, these results suggest there is substantial variance in RPA left unexplained when only using these selected field assessments.

The RSA₁₈₀ can be described as a RHIE assessment, with four high intensity actions performed over a brief period (∼7 s). The deceleration and change of direction actions in repeated shuttle performance presents a higher physiological load than that experienced in repeated effort straight line sprinting (Buchheit et al., 2010). The physiological and biomechanical demands between the RSA₁₈₀ and the LCMJ20 may have similarities based on the intensity of effort, the total duration of effort and the mechanical demands of both tasks. Specifically, the RSA₁₈₀ and the LCMJ20 are maximal intensity assessments that require high braking and propulsive forces to execute both the deceleration and acceleration and landing and propulsive elements of each assessment, respectively (Dos’Santos et al., 2017; Lake et al., 2021).

The stretch shortening cycle and corresponding elastic contribution of the muscle tendon units of the lower limb would also appear important for both assessments (Seiberl et al., 2021). Total duration of maximal effort is likely to be similar between both assessments. Total time of effort in the RSA₁₈₀ was 42.3 ± 1.5 s and when accounting for the braking, propulsive and landing phases of each jump, total time in the LCMJ20 is likely to be ∼35 s, with time per jump ranging from 1.5 s in the earlier repetitions (Lake et al., 2021; Pérez-Castilla et al., 2021) and, in visually inspecting the force-time data in this study, up to ∼2 s in later repetitions.

Despite the cyclical action of running and the acyclical action of the loaded CMJs, very strong relationships exist between the assessments due to these physiological and biomechanical similarities between the tasks. The strong association between the RSA₁₈₀ and the LCMJ20 RPA assessment may suggest that LCMJ20 is a good assessment of the underpinning physical quality(s) for high level RHIE performance.

HS_rel had a very strong negative relationship with LCMJ20, indicating that higher relative squat performance is associated with lower power decrement and therefore better RPA performance. As predicted 1RM loads were used to quantify external loads for the LCMJ20, participants with higher relative squat performance used higher absolute loads during the LCMJ20. Greater levels of HS strength improvement have previously been shown to enhance repeated high intensity performance (Bogdanis et al., 2011). Participants with greater HS strength performed better in the second half of a repeated sprint cycle assessment consisting of 10 × 6 s sprints interspersed with 24 s of passive recovery (Bogdanis et al., 2011). A similar finding can be observed in the current study with a strong negative relationship found between HS_rel and RSA₁₈₀.

It is suggested that stronger athletes may have better fatigue resistance through improved recruitment patterns and the ability to increase motor unit activity at the end of a repeated sprint test (Bogdanis et al., 2011). Similar neural adaptations may influence fatigue resistance in the LCMJ20. Additionally, both the externally applied load and the similar movement patterns used in the HS and the LCMJ20 are also likely to influence the very strong associations between the measures in comparison to the other assessments used in this study. Despite the strong associations between HS_rel and LCMJ20, the importance of HS_rel in predicting RPA performance is questionable. The inclusion of HS_rel to the predictive model only marginally improved the model. The lack of effect of HS_rel on the model may be due to the differences in the duration of both activities.

The inclusion of IRT2m in the predictive model also made little difference; with the model’s predictive ability slightly decreasing. The IRT2 is an intermittent field test with strong relationships to aerobic performance (Bangsbo, Iaia & Krustrup, 2008). The aerobic system has been found to be a key component of recovery between repeated sprints with the replenishment of phosphocreatine driven by aerobic processes (Turner & Stewart, 2013). Unlike repeated sprint performance, the LCMJ20 performance and the ability to repeat power does not seem to be highly determined by aerobic processes. This may reflect the short duration of each jump and recovery period being too brief for the aerobic system to influence recovery. In comparison to the IRT2, the intensity of effort in the LCMJ20 is maximal with a relatively short duration of effort. Anaerobic processes would therefore largely determine LCMJ20 performance, and this is evident in the strong relationship found between the Wingate assessment and other RPA assessments, with relationship ranging between r = 0.70–0.89 (Bosco, Luhtanen & Komi, 1983; Sands et al., 2004; Pupo et al., 2013; Fry et al., 2014).

Despite the very strong relationship found between RSA₁₈₀ and LCMJ20, there is substantial variance still left unexplained. With previous research identifying strong relationships between Wingate cycling performance and RPA assessments (Bosco, Luhtanen & Komi, 1983; Sands et al., 2004; Fry et al., 2014), a more direct measure of anaerobic capacity may have improved the predictive model in this study. However, in this study cardiorespiratory characteristics were examined using mechanically matched field assessments, all requiring running and the concomitant use of the stretch-shortening cycle. The LCMJ20 differs to several other RPA assessments reported in the literature (Bosco, Luhtanen & Komi, 1983; Sands et al., 2004; Fry et al., 2014), as it requires the use of the stretch-shortening cycle, a greater requirement for acceleration throughout the propulsive part of the movement and as a result of using external load, an increased mechanical demand placed on the propulsive and landing phases of the jump (Lake et al., 2021; Natera et al., 2023). The effect of additional load on landing phase force-time characteristics in the LCMJ20 is likely to further differentiate the Wingate assessment from the LCMJ20.

Unlike HS_rel, maximal outputs of lower limb force, lower limb power and speed did not improve the predictive model. These findings suggest that RPA, as measured by the LCMJ20, is not determined by maximal neuromuscular outputs. It is important to note that the force-time analysis used in the current study relies on discrete variables, taken from a single point in a jump, to explain jump performance. With the LCMJ20 used to assess the magnitude of decline in RPA performance, a more nuanced analysis methods to capture, detect and describe the differences in the whole force-time trace may be required to better describe RPA performance.

Although this study sought to describe RPA through commonly used field assessments, a limitation of this study was to have not included more sophisticated measures of both neuromuscular and cardiorespiratory function to help better explain and predict RPA. For example, surface electromyography may have been used to better understand changes in myoelectrical output of the lower limb muscles during the LCMJ20 (Felici & Del Vecchio, 2020). Near-infrared spectroscopy could also have been used to better understand various physiological and metabolic factors associated with LCMJ20 performance (Tuesta et al., 2022). Metabolic cart derived metrics, that are commonly utilised with aerobic and anaerobic ergometry based assessments, may have also provided additional variables that could have improved the predictive ability of our analyses. Considering the relationships between the 30 s Wingate assessments and other RPA assessments, the non-use of the Wingate assessment in this study may also be considered a limitation. Lastly, the sample size used in this study should be considered in any interpretation of the results.

Futures lines of investigations may look to include more sophisticated measures of neuromuscular and cardiorespiratory function to attempt to explain more of the variance in RPA and ultimately provide a more complete understanding of the factors contributing to RPA. Further investigations may also look to apply a more nuanced analysis of the whole LCMJ20 force-time waveform to gain greater insight into what exact changes occur across the 20 repeated CMJs. Lastly, further investigations may also look to develop RPA and assess its influence on RHIE performance through specific training methods like high volume power training (HVPT) (Apanukul, Suwannathada & Intiraporn, 2015; Gonzalo-Skok et al., 2016; Natera, Cardinale & Keogh, 2020; Schuster et al., 2018).

Despite the limitations of this study the findings suggest that RPA is in part a unique physical quality with substantial unexplained variance from the field assessments used. With RSA₁₈₀ having the strongest associations, and the only variable in the Stepwise predictor model to explain LCMJ20, RPA may well be an important physical quality that underpins RHIE performance. However, interpretation of the results of this study should be taken with care when applying to different participant groups or athletic cohorts. It is important to acknowledge that team sport athletes such as the field hockey players in this study, require a good combination of neuromuscular and cardiorespiratory development whereas a sprinter requires a highly developed neuromuscular system and an endurance athlete a highly developed cardiorespiratory system (Thompson, 2017; Lombard et al., 2021). Different findings to the current study might occur with athletes with more highly developed physical characteristics unique to their sports.

Conclusion

In attempting to identify the underlying neuromuscular and cardiorespiratory qualities that determine RPA, we found that approximately half of the variance in RPA can be explained by the common field assessments used in this study. With RPA having strong associations and strong predictive ability by RSA₁₈₀, we propose that RPA may be an underpinning physical quality that can be used to partly determine and assess RHIE performance. Other field assessments used in this study did not seem to associate with RPA performance and this may further highlight the potential uniqueness of RPA as a physical quality.

The LCMJ20 is an assessment of RPA that can be used in the field by practitioners and coaches to provide insight into the ability of their athletes to maintain maximal power output. With RHIEs having a strong relationship and predictive ability of RPA, the development of RPA may improve RHIEs in team sports like field hockey. Results of this study suggest that RPA is a somewhat unique physical quality that can underpin RHIE performance but is not associated with other common field-assessed physical qualities of fitness and strength. Practitioners can assess RPA and then determine whether to use specific training methods like HVPT to develop and enhance RPA and RHIE performance.

Supplemental Information