Abstract
The aim of this study was to assess the vertical jump performance and the force-velocity profile of elite female handball and volleyball players. Forty-one female athletes were measured, 28 handball players (age: 24.0 ± 3.6 years, body height: 1.75 ± 0.05 m, body mass: 69.0 ± 7.3 kg) and 13 volleyball players (age: 24.1 ± 5.2 years, body height: 1.83 ± 0.07 m and body mass: 74.9 ± 7.9 kg). All players performed unloaded and loaded countermovement jumps (CMJ) on a force platform. The theoretical maximal force (F0), the theoretical maximum velocity (v0), the theoretical maximal power (Pmax), the slope of the F-v relationship (Sfv) and the force-velocity imbalance (FVimb) were calculated. Mean value of vertical jump height was 0.33 ± 0.03m, with no difference between handball and volleyball players. Mean values of F0, v0, Pmax, Sfv and FVimb for all players were 31.2 ± 2.6 N/kg, 3.10 ± 0.50 m·s−1, 24.2 ± 3.2 w/kg, -10.32 ± 2.09 Ns/m/kg and 28.1 ± 13.3% respectively. Two players had a low magnitude velocity-deficit, whereas most of the players exhibited a low to high force-deficit. A strong correlation was found between the ratio of measured to optimal F-v slope with the change in the proportion of net force to total force during unloaded and loaded conditions. The findings suggest that it would be beneficial for these athletes to first decrease their force deficit through mainly maximal strength training before implementing training to further maximize power output. Establishment of the F-v profile could be a useful diagnostic tool for coaches to optimize strength training and to design training intervention based on the individual need of each athlete.
Introduction
Vertical jumping and the ability to generate high mechanical power during ballistic movements are key elements in volleyball and handball. In volleyball a higher jump during block jumps or during attacking strikes increases the probability for a successful action. 1 In handball a higher jump during a jump throw may allow for a shot over the block of the defence or may provide more time in the air and thus more options for an effective shot. 2 Testing and training of mechanical power is a usual practice in team sports3–7 with the countermovement jump (CMJ) being one of the most often used test protocols for lower limbs in routine performance measurements.8,9
Establishment of Force-velocity (F-v) profile during vertical jumping has been used to provide useful diagnostic information about the mechanical properties of the lower limbs.10,11 The F-v profile represents the relationship of the external force and velocity maximal capabilities. 10 Experimental evidence supports that each athlete demonstrates an optimal F-v relationship, which maximises performance, whereas a suboptimal relationship between force and velocity results in an imbalance in F-v profile (FVimb), which has been related with lower performance.10–12 Measurement protocols typically use unloaded and loaded vertical jumps and involve the quantification of the main variables of the lower limbs’ mechanical properties, namely the theoretical maximal force (F0), the theoretical maximum velocity (v0), the theoretical maximal power (Pmax) and the slope of the F-v relationship. The direction and the magnitude of the FVimb (the relative difference between the measured and the optimal slope of the F-v relationship) can be used by coaches and athletes to optimize external load in strength training by giving a more clear view of which strength qualities should be emphasized for each athlete individually. 12 In athletes with notable imbalance between force and velocity, it is suggested to first target the decrease of the FVimb before implementing training methods to increase maximal power output. 12
Changes in the F-v profile are indicative of specific strength training adaptations. 13 For example, heavy resistance training leads to changes primarily in the maximal force end, whereas ballistic training in the maximal velocity end of the F-v spectrum resulting in force oriented or velocity oriented profiles respectively.11,14 F-v profile also may distinguish between athletes from different sports15,16 and/or level of practice.15,16
Most F-v profile studies have involved male athletes. Assessment of F-v profiles with female athletes is limited in the literature; in most studies with female athletes, the authors used squat jump (SJ) and not the countermovement jump (CMJ).15–17 However, the dynamic conditions of CMJs are more common in many sport activities (e.g. in handball and volleyball) involving successive eccentric and concentric muscle contractions, than the concentric only action of SJ. Assessment of the F-v profile in elite female athletes could provide a deeper insight into the maximal mechanical properties of the lower limbs during the countermovement jump allowing to design training interventions based on the individual force velocity relationship. Also, it may provide a reference in the evaluation of F-v profile in female athletes during CMJ.
Therefore, the primary purpose of this study was to assess the force-velocity profile of elite female handball and volleyball players during the countermovement jump. It was assumed that the F-v profile of the athletes would differ due to differences in jumping movement patterns (bilateral jumps from a stationary position occur more frequently in volleyball, than in handball) and in the athletes’ training periodization. A secondary purpose was to examine the relationship between percent changes in ground reaction force (GRF) and in resultant force under loaded and unloaded conditions with the FVimb. It was hypothesized that the magnitude of the percent changes in GRF and resultant force would correlate with the force-velocity imbalance. While the GRF is expected to increase as the external load increases, the resultant force shows a progressive decrease resulting in a significant decrease in average velocity 18 and eventually in power output. 19 The proportional changes in GRF and resultant force under various loading conditions may provide a better understanding of the athletes’ strength qualities.
Materials and methods
Participants
Forty-one female athletes were measured (28 handball and 13 volleyball players). All participants were members of the respective Hungarian Women’s National Team participating in international competitions. Athletes who reported any injury, pain or illness at the time of the measurements were excluded. Five players were excluded from the study, three of them did not complete all unloaded and loaded jumps, while two players had vertical jump height below 8 centimetres during the loaded jumps. Therefore, the results of 36 players were used in the statistical analysis. At the time of the measurements the handball players were at the beginning of their regular in-season phase (early September), while the volleyball players had just completed their regular season and were at the beginning of a preparation period with the national team (late April, early May). A typical training week of the players included five sessions of sport specific training (120 min/session) and two sessions of strength training of about 60-75min/session. All athletes were informed about the types of the measurements and gave their written consent to perform the measurements. This study was approved by the University’s Research Ethics Committee (approval number: TE-KEB/3/2019). Descriptive characteristics of the participants are presented in Table 1.
Characteristics of the subjects (mean±SD).
BMI: body mass index.
*p < 0.05 between handball and volleyball players.
Procedures
Athletes arrived at the laboratory in the morning hours (8.00–8.30 a.m.). Body height (BH) was measured with an anthropometer (DKSH Management, Zürich, Switzerland) to the nearest 0.1 cm. Body mass (BM) and body composition was measured by bioimpedance method (InBody720, Biospace, Seoul, Korea), participants were asked to refrain from consuming foods and fluids before this measurement.
Vertical jump was measured on a 0.6 × 0.6 m force platform (FP4, HUR Labs Oy, Tampere, Finland). Sampling rate was set at 500 Hz. Prior to the testing the participants performed a warmup, which included easy running, skipping drills, and dynamic stretching exercises. Verbal instructions and familiarization with five submaximal jumps were also given. Then the athletes performed maximal unloaded jumps without an arm swing, with hands kept on hips and maximal jumps against additional loads of 25%, 50%, 75% and 100% of their body mass with hands kept on the bar. Loaded jumps were performed on a smith machine (Atlas-sport Ltd.) that allowed vertical displacement of the bar through a fixed path. Three attempts were performed for unloaded jumps and two attempts for loaded jumps with two minutes of passive rest between each attempt and 3-4 minutes between load conditions. The best values for each jump type were recorded for statistical analysis. Verbal motivation and encouragement were given to maximize effort. CMJs were performed after a one second quite standing on the force platform, then the athletes bent their knees to approximately ninety degrees and then immediately jumped as high as possible. A 2.5SD threshold was used to determine the initiation of the force-time curve. All data were collected and analysed with HUR Labs Force Platform Software Suite 2.40 (HUR Labs Oy, Tampere, Finland).
Average force for each loading condition was extracted from the force platform during the push-off phase (from the moment ground reaction force exceeds body weight to the instant of takeoff). Total ground reaction force (GRF) was calculated as the force due to the system’s weight (= body weight + external load), whereas net (or resultant) force was calculated as the force above the system’s weight (= total GRF – system weight). The proportion of net force to GRF (net F:GRF) was calculated for each loading condition. Changes (in %) in net F:GRF were calculated as the difference between the unloaded jump and the loaded jump against 100% of body mass. Jump height was obtained from force platform data based on the impulse-momentum method. 20 Impulse during the unloaded jumps was calculated as the product of resultant force (GRF – body weight) multiplied by the ground contact phase (from the moment GRF exceeds body weight to the moment it drops to become equal to body weight) and it was extracted from the force platform software. Mean velocity was calculated from jump height as proposed previously. 10 F-v relationship was determined using least squared linear regressions. Normalized to body mass force and velocity data were extrapolated to obtain F0 (theoretical maximum force at null velocity) and v0 (maximum velocity with no load). Theoretical maximum power was calculated as Pmax= F0 · v0/4.10–12 Theoretical optimal F-v profile maximizing jumping performance was calculated from Pmax and push-off distance (hpo).10,11 Push-off distance was defined as the vertical displacement of the centre of mass (COM) from the lowest point of the countermovement to the instant of takeoff and was extracted from the vertical displacement-time curve of the force platform. Finally, individual F-v imbalance (FVimb) for each player was calculated as the difference (in %) between measured and optimal F-v slope (normalized to body mass in N·s·Kg−1·m−1). 11 Based on previously reported evaluation of the FVimb 14 the following categories were defined: well-balanced (0 ± 10%), low deficit (10–40%), high deficit (>40%).
Statistical analysis
Values are expressed as means ± SD. Reliability was measured using absolute agreement 2-way mixed effects model intraclass correlation coefficient (ICC). Kolmogorov-Smirnov test was used to assess normality. Independent sample t-test and standardized effect sizes (Hedge’s g with correction for small samples) was used to compare differences between handball and volleyball players. Magnitude of the difference to evaluate the between-groups effect sizes was: 21 trivial (<0.2), small (≥0.2), moderate (≥0.5) and large (≥0.8). Pearson’s correlation was used to quantify the relationship between the ratio of measured to the optimal F-v slope with the percent change of net F:GRF between the unloaded jump and the loaded jump against 100% of body mass. Statistica 13.5 for windows (TIBCO software Inc, Palo Alto, CA, USA) statistical package was used for the statistical analysis.
Results
ICC for jump height was 0.91 (95% CI: 0.85-0.95), for average force 0.95 (95% CI: 0.91-0.97) and for maximum net impulse 0.99 (95% CI: 0.98-0.99). Volleyball players were taller (t=-3.88; p < 0.05) and heavier (t=-2.24; p < 0.05) than handball players and had greater lean body mass (t=−2.40; p < 0.05) (Table 1). There was no difference between the two groups in jump height, average force, and time to takeoff. A large effect difference was found only for maximum impulse; volleyball players had higher values than handball players (Table 2).
Descriptive results of the unloaded vertical jump (mean±SD).
*p < 0.05 between handball and volleyball players.
11.1% (4 players) of the sample showed a well-balanced F-v profile (within ±10% from optimal), 13.9% (5 players) presented a high deficit (>40% from optimal), whereas 75% (27 players) a low deficit (from 10 to 40% from optimal) in F-v profile. Volleyball players demonstrated larger pushoff distance and a slightly smaller F-v imbalance than handball players (Table 3). Loss in performance due to FVimb was about 5% (2.5 cm), which was even more than 10 cm for the athletes with the most unfavourable F-v profile. Athletes with high force-deficit (FVimb >40%) had a loss in performance of 4 cm or more.
Descriptive results of the F-v profile (mean±SD).
F0: theoretical maximum force; v0: theoretical maximum velocity; Pmax: theoretical maximum power; hpo: push-off distance; Sfv_act: slope of measured force-velocity relationship; Sfv_opt: slope of optimal force-velocity relationship; FVimb: force-velocity imbalance; hmax: theoretical maximum jump height at optimal force-velocity relationship; JH:hmax: ratio of the actual and the theoretical maximum jump height; JH deficit: loss in performance due to force-velocity imbalance.
*p < 0.05 between handball and volleyball players.
The proportion of net F:GRF decreased progressively during loaded jumps, differences between each loading condition were significant (F4,175=190.6; partial η2=0.81; p < 0.01), except between loaded jump with 75%BM and 100%BM (Figure 1). The decrease in net F:GRF between unloaded jump and loaded jump with 100%BM ranged from 15.8% to 30.3% with a mean±SD decrease of 24.4 ± 3.4%. There was a large correlation (r=−0.70; p < 0.01) of the ratio of measured to the optimal F-v slope with the decrease in Fnet:FGRF between unloaded jump and loaded jump with 100%BM (Figure 2).

Proportion (in%) of net force to total GRF for each loading condition.

Relationship (95% CI) of the ratio of measured force-velocity slope to the optimal force-velocity slope with the change (%) in the proportion of net force to total GRF between unloaded jump and jump with additional load of 100% of body mass.
Discussion
The main purpose of this study was to assess force-velocity profile of elite female handball and volleyball players during CMJ. The main concept in establishing the athletes’ F-v profile is that strength and power training based on the individual mechanical characteristics has been suggested to improve training effectiveness and performance in ballistic movements. 12
The results revealed low to high force-deficit in most of the players. About 10% of the players had near optimal F-v relationship (±10%), and only two players had velocity-deficit of low magnitude. These results are comparable with previous findings.11,17,22 In male athletes (soccer players, sprinters, rugby players) Samozino et al. (2014) reported imbalance of mainly force-deficit (mean±SD FVimb: 42.6 ± 34.4%). In track and field male athletes Jimenez-Reyes et al. (2014) reported mean FVimb values of 43.7 ± 16.1%. In a study with female soccer players, using, however SJ and not CMJ, significant force deficit was found (mean±SD FVimb: 64.5 ± 16.3%). 17
The remarkably larger prevalence of force-deficit observed here most probably reflects the type of strength training the athletes were used to. In general, athletes, who used to train against high resistance loads usually show better developed force qualities and therefore force oriented F-v profile.11,14 Based on personal communications with the strength and conditioning coaches, the frequency of the athletes’ strength training program was one or two times per week and typically included bodyweight exercises or use of submaximal loads (60–75% 1RM), which allow repetitions from 8 to 15. This type of mostly hypertrophic based strength training, however, seems not effective enough to improve maximal strength and maximize power output under loaded conditions.
Force deficit emphasizes the need to improve maximal strength qualities. Based on our results we might argue that the athletes in our sample do not possess a satisfying level of maximal strength; force production at low velocities (high resistance load) was smaller than expected (according to their optimal F-v slope). This suggests that it would be beneficial for these athletes to first decrease their force deficit before implementing training to further maximize power output.12,14 Strength training in this case should focus primarily on the force side of the F-v spectrum, applying high resistance loads for the high force-deficit group and strength-power training with an emphasis on strength (∼80-85%) for the low force-deficit group. Promising results have been reported when training was designed based on the individual F-v profile. Jimenez-Reyes et al. (2017) have confirmed experimentally that strength training designed according to the individual F-v profile can reduce FVimb and consequently increase performance in vertical jumping. Athletes with initial force-deficit in their F-v profile, who followed heavy-load resistance training had significant improvements in both F0 and FVimb. 14 Moreover, when the training program was individualized based not only on the initial FVimb, but also on the duration of the program the effectiveness of the training was even higher. 14
In another study involving different sports and levels of practice Jimenez-Reyes et al. (2018) found slight differences in the mechanical properties between female elite level handball and medium level volleyball players assuming different F-v profile. Handball players had mostly force-oriented profiles, while volleyball players had velocity-oriented profiles. 16 Performance in SJ was similar, however, considering their level of practice (elite vs. medium respectively) we assume a higher performance in vertical jump for volleyball players. Conversely, our findings revealed only small (not significant) difference in jump height and no difference in the mechanical qualities. Moderate difference in v0 was even at opposite direction than those reported in the study of Jimenez-Reyes et al. (2018). According to our expectations and based on sport specific patterns (bilateral vertical jumps from a stationary position are more typical in volleyball, than in handball) volleyball players would demonstrate more developed jumping abilities, however, this was not confirmed by the results. Furthermore, the differences observed in the push-off distance indicate more advanced ability to express high mechanical power output for handball players. Being similar in force production, small push-off distance (handball players) may underestimate, while large push-off distance (volleyball players) may overestimate the power capability of an athlete. 23
Push-off distance also affected the differences in F-v profile (Sfv_opt and FVimb). Push-off distance is used in the determination of the optimal F-v relationship, its magnitude has a direct effect on the F-v imbalance. A small hpo increases the steepness of the optimal F-v slope, which in turn may increase the difference between the measured and the optimal F-v slope, resulting in larger FVimb. Conversely, larger hpo decreases the steepness of the optimal F-v slope, which results in less imbalanced F-v profile. This was the case for the volleyball players, who had a more balanced F-v profile, explained however, with their larger hpo.
There is no clear explanation for the lack of distinct differences in the F-v profile between the two sports. Volleyball players were at the beginning of a 4-month preparation period with the national team, being under large training load at the time of the measurements and performing mainly hypertrophic based strength training. This partly could explain their lower average velocity values, since this type of training limits velocity maximization and high power production. Their type of training together with the accumulated fatigue may had influenced their jumping performance and their F-v profile. Handball players were at the beginning of the competitive period (assuming power developing training) and thus we would expect a force-oriented profile with higher Pmax values. In power training it is recommended to apply various resistance loads, which affect the entire range of the force-velocity spectrum optimizing in this way power development and eventually enhancing athletic performance. 13 Ideally, this should be divided in three training periods based on the 3-step model 24 starting with the increase in the muscle cross-sectional area followed by maximal strength and then by power development. 25 Irrespective of the different training period the athletes were at the time of the measurements both groups followed a similar strength training methodology and had a similar vertical jump performance and a F-v profile. These similarities may indicate a less accurate approach in strength training and periodization, mainly with regards to the maximal strength and power development periods.
An interesting aspect in loaded vertical jumping is the percent changes in force production expressed by the proportion of net force to total GRF. During unloaded jumps this proportion is about 45-50%, and is a determinant factor in achieving a high jump height. 20 As external load increases during loaded jumps total average force also increases, whereas net average force decreases resulting in a reduction in the proportion of net force to total force (Figure 1). In loaded jumps with additional load of 100%BM only about one fourth of total force is attributed to net force. This means, that with the increase of the external load, the force applied by the athlete is used mostly to overcome the increased total mass, rather than accelerate it. 18 A high relationship was found between changes in force proportion with the force-velocity profile (Figure 2). Large decrease in this proportion is associated with F-v profile of force-deficit, resulting in larger FVimb. In fact, a favourable F-v profile would incorporate large increase in total force and small (or none) decrease in net force. This suggests that as external load increases the athlete is still able not only to overcome the increased inertia, but also to accelerate the system (body weight+external load), which assumes well developed strength capabilities. Athletes with a decrease in the proportion of net force to total force less than 20-25% had comparatively a well-balanced F-v profile. When this decrease was more than 25% a force-deficit was observed, indicating the need for further strength development. Percent changes in the proportion of net force to total GRF may offer an additional tool in the evaluation of the athletes’ strength capabilities during loaded jumps.
A main limitation of this study is the absence of the exact training data, therefore straightforward conclusions regarding the F-v profile of the athletes cannot be made, nor can these results be generalized or transferred to other athletes. It should be noted, however, that the athletes in our study were members of various clubs, thus following different training programs, which complicated collection of training data. Another limitation is the variations within a microcycle. The athletes appeared in the laboratory in different days of the week, thus training or rest days preceded the measurements may also have had influenced their F-v profile. Finally, there are no data concerning the level of the athletes’ maximal strength, therefore we cannot relate maximal strength with the F-v profile. Future research should focus on determining levels of maximal strength, which optimizes the F-v profile in elite female athletes.
Conclusions
In this study we described the F-v profile of elite female handball and volleyball players. The results revealed low to high force-deficit in most of the players highlighting the need to decrease force-velocity imbalance mainly through maximal strength training. Findings also provide reference data for the evaluation of the F-v profile during the CMJ in elite female athletes. F-v profile data could be of practical importance for coaches to describe more accurately the mechanical qualities of their athletes and to design training methods optimizing strength and power development.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
