Biomechanical Profile of Portuguese High-Level Female Handball Players

Abstract

Background/Objectives: This study aimed to investigate the anthropometric characteristics, motor performance, and isokinetic strength profiles of elite Portuguese female handball players, as well as to examine the relationships among these variables. Methods: Sixteen national-team female handball players with an average age of 20.25 ± 0.45 years, height of 171.13 ± 8.13 cm and body mass of 72.24 ± 10.96 kg volunteered. Evaluations were conducted in two sessions within one week (24–48 h apart). The first comprised anthropometric and motor performance tests, while the second focused on isokinetic strength assessments of the upper and lower limbs. Pearson correlations assessed variable associations (p < 0.05). Results: Direct correlations were found between height and arm span (r = 0.910) and between internal rotation total work and internal rotation average power (r = 0.960). The 9 m jump throw was associated with the 7 m standing throw (r = 0.670). External rotation peak torque correlated with squat jump performance (r = 0.540) and the 7 m standing throw (r = 0.760) and 9 m jump throw (r = 0.568). Internal rotation peak torque associated with squat jump performance (r = 0.674) and the 7 m standing throw (r = 0.550). Knee extension peak torque correlated with squat jump performance (r = 0.650), while knee extension total work was strongly associated with external rotation total work (r = 0.870). Knee flexion total work was associated with knee flexion peak torque (r = 0.910). Conclusions: The integrated analysis of anthropometric, motor and isokinetic variables revealed distinct strength–performance associations in female handball players, highlighting the role of upper- and lower-limb muscle function in jumping and throwing.

1. Introduction

Female handball is played at a high-intensity intermittent pace with repeated explosive actions separated by short recovery periods, demanding contributions from both anaerobic and aerobic energy systems while requiring high levels of strength and power [1,2]. Players perform sprints of 5–30 m, rapid directional changes, accelerations, decelerations and vertical or horizontal jumps during training and games [3]. They also execute various throwing techniques (jump, standing and running throws), often under defensive pressure within complex tactical contexts [4]. Training loads may consist of 5–7 weekly sessions focused on strength, power and plyometric work, with a high volume of jumps and throws accumulated across the competitive season, which impose substantial mechanical stress on the musculoskeletal system [5]. Such chronic load promotes neuromuscular and metabolic adaptations, enhancing rapid force production, fatigue resistance and motor control, which are crucial for maintaining performance during decisive game phases [5].

Building on these physical and technical demands, optimal body composition and anthropometric profiles, particularly sufficient lean mass, reduced body fat, greater stature and extended arm span are crucial for sustaining high-level performance in female handball [6]. Vertical jump performance, assessed through countermovement and squat jumps, generally reaches heights of 0.29 m, while horizontal jumps average around ~1.9 m. Throwing velocities usually reach ~80 km/h from 7 m and ~83 km/h from 9 m, reflecting the high-power demands of the sport [3]. These performance metrics provide essential benchmarks for systematic monitoring using accessible and reliable tests to establish functional profiles, guide individualized training strategies and potentially mitigate the biomechanical demands that contribute to injury occurrence in female players [7].

Given the high-intensity and repetitive nature of female handball, the shoulder and knee joints are particularly vulnerable to injury. In handball players, a significant proportion report shoulder pain, estimated at 19–36% per season, and this is primarily due to overuse from repetitive throwing actions [8]. Knee injuries are also common, with ~7–27% of players sustaining anterior cruciate ligament ruptures during their careers and meniscal injuries occurring in more than 16% of cases [9]. Such injuries frequently result from high braking forces, torsional loading, rapid change in direction and landing tasks, stressing the need for early implementation of structured prevention programs. Evidence demonstrates that such interventions can reduce injury risk by approximately half, supporting their integration into training routines of elite female players [10].

To address these vulnerabilities and better understand potential performance-limiting imbalances, isokinetic evaluation provides objective measures of muscle function under controlled conditions. In female handball players, shoulder internal rotation strength was significantly greater than external rotation strength, reflecting the functional adaptations of repeated high intensity [11]. At the knee, extensor strength exceeded flexor strength by ~70–75%, corresponding to a hamstring-to-quadriceps ratio of around 0.58–60°·s⁻¹, reflecting adaptations to the functional demands on lower-limb power in elite female handball players [12]. These findings provide a robust basis for identifying asymmetries that may predispose players to injury and for prescribing targeted strength programs to optimize muscle function and reduce injury risk in elite players [12].

Although anthropometric characteristics, motor performance and isokinetic strength variables have each been studied individually in female handball, few investigations have examined their integrated relationships, despite existing evidence on the sport’s physical demands, performance metrics and injury susceptibility. Most of the available research isolates aspects such as isokinetic torque or anthropometry in sub-elite or youth categories, often reporting only weak-to-moderate correlations [13,14,15]. This lack of integrated analysis limits the translation of findings into training and rehabilitation programs that reflect the real physical and technical demands of national-team players. To address this gap, the current study offers a comprehensive analysis that integrates anthropometric, motor performance, and isokinetic strength variables in elite Portuguese female handball players. This integrated approach provides a more complete physical profile to inform evidence-based strategies for training, injury prevention, and rehabilitation in high-performance settings. By identifying key associations, the findings may not only help athletes optimize performance but also support physiotherapists, coaches, and players in developing training programs, rehabilitation protocols, and injury-prevention interventions tailored to the specific demands of elite female handball.

Accordingly, the current study aimed to characterize and examine the relationships between anthropometric characteristics, motor performance and isokinetic strength of the upper and lower limbs in elite female players from the Portuguese national handball team. It was hypothesized that (i) the players anthropometric and performance characteristics would be consistent with normative values reported for elite female handball, and (ii) isokinetic measures of the shoulder and knee would be positively correlated with jump and throwing performance, with anthropometric variables (e.g., arm span, lean mass and height being associated with motor and muscle strength capacities).

2. Materials and Methods

2.1. Participants

Sixteen high-level female handball players, all healthy and without injuries in the six months prior to testing (competing at both national and international levels) and currently representing the Portuguese national team, voluntarily participated in the current study. The evaluations were conducted in two sessions within a one-week period, with a minimum interval of 24–48 h. The first session included anthropometric measurements and general motor performance assessments, while the second focused on isokinetic testing of upper and lower limb muscle function. All players were fully informed about the experimental procedures and potential risks associated with their participation, and each provided written informed consent before the study commenced. The study procedures were approved by the Ethics Committee (CEFADE 27/2020) in accordance with the Declaration of Helsinki and the guidelines of the World Medical Association for research involving human participants.

Before each testing session, all players completed a standardized 10 min warm-up consisting of dynamic mobility and low-intensity aerobic exercises targeting the major joints and muscle groups. The first session included anthropometric assessments followed by motor performance tests in the following order: (1) countermovement and squat jumps, (2) horizontal jump, (3) throwing velocity (standing, jump, and three-step throws), and (4) handgrip strength. The second session comprised isokinetic assessments in a fixed order, starting with shoulder internal/external rotation, followed by knee flexion/extension. A minimum rest interval of 3–5 min was provided between tests to reduce fatigue. To ensure consistency, each station was administered by the same experienced assessor throughout the procedures.

2.2. Experimental Methods and Procedures

Regarding the anthropometric measurements, height was assessed using a Seca 206 stadiometer (Seca GmbH, Hamburg, Germany) with a precision to the nearest millimeter, while arm span was measured with an anthropometer and sliding caliper (Siber-Hegner, GPM, Zurich, Switzerland). Body mass, body fat and fat-free mass were determined using a bioelectrical impedance scale (InBody 230, Biospace, Seoul, Republic of Korea). The technical error of measurement was 0.8 cm for height, 0.9 cm for arm span, 0.1 cm for hand breadth and length, and 0.9 kg and 0.3 kg for body mass and fat-free mass, respectively. All assessments were conducted in accordance with the International Working Group on Kinanthropometry protocols [16].

General motor performance was assessed using a battery of standardized tests and calibrated equipment. Lower-limb explosive power was measured through countermovement and squat jumps performed on a contact platform (Bertec Inc., Columbus, OH, USA; 60 × 90 cm; 2000 Hz). Three maximal trials were performed for each jump type, with a rest interval of at least 30 s between attempts. Jump height was calculated using the flight-time method and the mean of the three attempts was retained for analysis [17]. Jump power was calculated as the product of ground reaction force and center-of-mass velocity (derived from the force signal), and the maximal power value was used for analysis. Outcomes were expressed as jump height (cm) and relative power (W·kg⁻¹) [18].

Horizontal jumping ability was assessed using the standing horizontal jump test, with participants starting behind a marked line. Each players performed three attempts, and the mean distance achieved was recorded in centimeters for analysis [3]. Throwing velocity was evaluated in three conditions: a standing throw from 7 m, a jump throw from 9 m and a standing throw after 3 steps. All throws were performed toward a standard handball goal without opposition and measured using a Stalker ATS II radar gun (Applied Concepts Inc., Stalker Radar, Plano, TX, USA) positioned on a tripod two meters behind the goal [16].

Handgrip strength was assessed using a digital hand-held dynamometer (model T.K.K.5401 Grip-D, Takei, Japan; accuracy ± 2.0 kg·f). Assessments were performed separately for the dominant and non-dominant hands, with participants standing upright and the arm fully extended along the side of the body, without flexion or rotation. Each participant completed three maximal trials per hand, with at least 30 s of rest between attempts to minimize fatigue. The highest value obtained for each hand was retained for analysis [19].

Shoulder rotator and knee flexor and extensor muscle strength were evaluated using an isokinetic dynamometer (Biodex Multi-Joint System 4, Biodex Medical System, Shirley, New York, NY, USA) [13,20]. All isokinetic strength values were obtained directly from the device reports. Peak torque was defined as the highest torque value achieved during each set of five repetitions. No gravity correction was performed in the shoulder analysis, while gravity correction was applied in the knee tests. Torque data were not filtered. For the upper limb, players were seated and stabilized according to the manufacturer’s recommendations to ensure that only the shoulder rotation movement was performed, with the dominant upper limb evaluated at an angular velocity of 60°·s⁻¹ (5 repetitions) [15]. The tested upper limb was positioned at 90° of shoulder abduction with the elbow flexed at 90° (90/90 position), allowing internal rotation to 50° and external rotation to 90°, resulting in a total range of motion of 140° (with the neutral position was defined as parallel to the ground) [21]. Regarding the dominant lower limb assessment, players were seated with the torso and pelvis securely stabilized. Players completed 5 maximal concentric knee extension and flexion repetitions at an angular velocity of 60°·s⁻¹, using a standardized range of motion from 90° of knee flexion to full extension (0°) [22]. Verbal encouragement was provided throughout all tests to ensure maximal effort. The assessment protocol is illustrated in Figure 1, detailing the procedures for anthropometric, motor performance, and muscle strength evaluation in handball players.

2.3. Statistical Analysis

An a priori sample size calculation using G*Power software (version 3.1.9.7; Heirich-Heine-Universität Düsseldorf, Germany), assuming α = 0.05, power = 0.80, and an expected large effect size (ρ = 0.50), indicated that 29 participants would be required. However, as the study involved a national team, all available athletes (n = 16) were included. All statistical analyses were performed using SPSS version 30.0 for Windows (IBM Corp., Armonk, NY, USA), with mean and standard deviation (SD) values were calculated and reported for all variables. The Levene and Shapiro–Wilk tests were applied to verify the assumptions of homogeneity of variances and normality of data distribution. Pearson’s correlation coefficient was used to examine the relationships between the players’ anthropometric characteristics, general motor performance variables and isokinetic strength measures, with correlation strength interpreted as small (0.1–0.3), medium (0.3–0.5) and large (0.5–1.0). For the most relevant correlations, 95% confident intervals (CI) were calculated, and the coefficient of determination (R²) was reported as an effect size representing the proportion of variance explained. The significance level was set at p < 0.05.

3. Results

The high-level handball players mean physical characteristics were as follows: 20.25 ± 0.45 years of age, body mass of 72.24 ± 10.96 kg, height of 171.13 ± 8.13 cm, arm span of 175.16 ± 10.98 cm, body mass index of 24.43 ± 3.10 kg·m⁻², skeletal muscle mass of 31.23 ± 4.28 kg, and fat mass of 17.79 ± 6.19 kg. Most players (75%) were right-hand dominant, while 25% were left-hand dominant.

Table 1. Mean and SD values of players’ general motor performance variables.

Variables	n = 16
Countermovement jump height (cm)	32.30 ± 4.30
Countermovement jump height (W/kg)	44.53 ± 4.94
Squat jump height (cm)	30.01 ± 4.33
Squat jump height (W/kg)	44.30 ± 5.21
Horizontal jump (cm)	196.81 ± 19.84
7 m standing throw (km/h)	76.56 ± 6.76
9 m jump throw (km/h)	83.29 ± 6.12
Standing throw after three steps (km/h)	83.43 ± 6.85
Right hand dynamometry (kg·f)	37.37 ± 4.61
Left hand dynamometry (kg·f)	37.20 ± 4.10

shows that players exhibited comparable vertical jump heights between squat and countermovement jumps (both in cm and W/kg), whereas horizontal jump distances averaged ~196 cm. Throwing velocities were greater in jump throws compared to standing throws from 7 m, with similar speeds observed between the jump throw and the standing throw after 3 steps. Handgrip strength values were generally balanced between the right and left hands, with slightly higher results for the right hand.

Table 2. Mean and SD values of isokinetic variables for shoulder external and internal rotation.

Variables	External Rotation	Internal Rotation
Peak torque (N·m)	28.64 ± 6.32	39.01 ± 8.93
Peak torque (% of body mass)	39.86 ± 7.45	54.35 ± 10.52
Average torque (N·m)	26.62 ± 5.81	35.34 ± 8.15
Total work (J)	258.66 ± 67.01	316.30 ± 70.31
Total work (% of body mass)	63.63 ± 23.64	79.43 ± 30.35
Average power (W)	21.68 ± 5.47	25.80 ± 80
Amplitude (°)	138.88 ± 1.05	__
External/internal rotation ratio (%)	74.17 ± 9.87	__

shows that internal shoulder rotation produced higher values than external rotation across all isokinetic variables. Peak torque was ~36% greater in internal rotation, both in absolute terms and relative to body mass. Average torque was about ~33% higher, while total work increased by roughly ~22% in absolute values and ~25% when adjusted for body mass. Average power in internal rotation exceeded external rotation by ~19%. The amplitude of motion for external rotation averaged ~139° and the external-to-internal rotation ratio was ~74%, indicating a balanced but lower relative contribution of external rotation when compared to internal rotation. Overall, internal rotation demonstrated consistently superior performance, with the largest differences observed in torque-related measures.

Table 3. Mean and SD values of isokinetic variables for knee extension and flexion.

Variables	Extension	Flexion
Peak torque (N·m)	182.78 ± 26.07	92.81 ± 17.49
Peak torque (% of body mass)	261.75 ± 41.08	133.85 ± 23.70
Average torque (N·m)	167.72 ± 27.66	85.09 ± 17.33
Total work (J)	792.63 ± 163.65	486.11 ± 143.66
Total work (% of body mass)	194.01 ± 87.39	116.87 ± 51.83
Average power (W)	122.80 ± 18.45	67.67 ± 13.93
Amplitude (°)	79.93 ± 8.44	__
Extension/flexion ratio (%)	50.99 ± 7.82	__

display that knee extension generated substantially greater values than knee flexion across all isokinetic parameters. Peak torque was nearly double in extension, showing increases of about ~97% in both absolute measurements and values normalized to body mass. Average torque displayed the same development, with extension outperforming flexion by ~97%. Total work was higher in extension by roughly ~63% in absolute terms and ~66% relative to body mass, while average power exceeded flexion by about ~81%. The range of motion recorded for extension was ~80° and the extension-to-flexion ratio reached ~51%, highlighting the markedly superior force production capacity in extension compared with flexion.

The associations between the assessed variables are presented in

Table 4. Pearson correlation coefficients between anthropometric, motor performance and isokinetic strength variables.

Variables	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19
Weight (kg) (1)
Height (cm) (2)	0.44
Span (cm) (3)	0.50	0.91 **
Countermovement jump height (W/kg) (4)	0.42	−0.06	0.02
Squat Jump (W/kg) (5)	−0.41	0.07	0.05	−0.05
7 m standing throw (km/h) (6)	0.28	0.18	0.22	0.31	0.38
9 m jump throw (km/h) (7)	0.13	0.20	0.18	−0.07	0.15	0.67 **
External rotation peak torque (% of BM) (8)	−0.05	0.06	0.07	0.03	0.540 *	0.76 **	0.568 *
External rotation total work (% of BM) (9)	−0.53	0.16	0.10	−0.23	0.541 *	0.08	0.19	0.50
External rotation average power (W) (10)	0.51	0.16	0.25	0.39	0.14	0.80 **	0.44	0.78 **	0.04
Amplitude (°) (11)	0.01	0.13	0.18	−0.23	0.34	0.27	0.45	0.25	0.08	0.11
Internal rotation peak torque (% of BM) (12)	−0.04	0.11	0.30	0.05	0.674 **	0.55 *	0.30	0.75 **	0.54 *	0.57 *	0.23
Internal rotation total work (% of BM) (13)	0.44	0.35	0.46	0.13	0.32	0.80 **	0.49	0.74 **	0.24	0.84 **	0.16	0.74 **
Internal rotation average power (W) (14)	0.54 *	0.32	0.48	0.32	0.27	0.78 **	0.44	0.71 **	0.14	0.88 **	0.14	0.76 **	0.96 **
Knee extension peak torque (% of BM) 15)	−0.34	0.24	0.20	0.21	0.65 *	0.55 *	0.31	0.54 *	0.63 *	0.30	0.16	0.47	0.38	0.31
Knee extension total work (% of BM) (16)	−0.65 *	0.21	0.08	−0.20	0.50	−0.11	0.04	0.14	0.87 **	−0.33	−0.02	0.23	−0.10	−0.19	0.65 *
Knee flexion peak torque (% of BM) (17)	−0.07	−0.07	−0.01	0.39	0.00	0.49	0.17	0.50	0.30	0.56 *	−0.18	0.24	0.38	0.37	0.63 *	0.17
Knee flexion total work (% of BM) (18)	−0.59 *	0.10	0.01	−0.10	0.29	−0.07	0.05	0.19	0.82 **	−0.17	−0.10	0.19	−0.08	−0.15	0.67 **	0.91 **	0.47
Knee extension amplitude (°) (19)	−0.59 *	−0.11	−0.19	0.06	0.45	0.05	0.15	0.33	0.61 *	−0.09	0.04	0.32	−0.17	−0.11	0.45	0.65 *	0.19	0.63 *

* p ≤ 0.05 and ** p ≤ 0.01. BM: body mass.

, with height exhibiting a large correlation with arm span (r = 0.91; 95% CI [0.78, 0.98]; R² = 0.82; large), while the 9 m jump throw showed a large correlation with the 7 m standing throw (r = 0.67; 95% CI [−0.005, 0.14]; R² = 0.45; large). External rotation peak torque displayed large correlations with squat jump performance (r = 0.54; 95% CI [0.11, 0.83]; R² = 0.29; large) and both the 7 m standing throw (r = 0.76; 95% CI [0.50, 0.93]; R² = 0.58; large) and 9 m jump throw (r = 0.57; 95% CI [0.16, 0.82]; R² = 0.32; large), whereas external rotation average power demonstrated large correlations with the 7 m standing throw (r = 0.80; 95% CI [0.52, 0.95]; R² = 0.64; large) and external rotation peak torque (r = 0.78; 95% CI [0.55, 0.93]; R² = 0.61; large). Internal rotation variables (peak torque, total work and average power) were strongly interrelated, with total work and average power presenting a large association (r = 0.96; 95% CI [0.89, 0.99]; R² = 0.92; large) and large associations with external rotation power (r = 0.84; 95% CI [0.66, 0.93]; R² = 0.42; large) and peak torque (r = 0.75; 95% CI [0.20, 0.97]; R² = 0.56; large). Knee extension peak torque correlated largely with squat jump performance (r = 0.65; 95% CI [0.32, 0.87]; R² = 0.42; large), while knee extension total work was largely associated with external rotation total work (r = 0.87; 95% CI [0.77, 0.96]; R² = 0.76; large) and exhibited a large negative correlation with body mass (r = −0.65; 95% CI [−0.84, −0.39]; R² = 0.42; large). Knee flexion total work showed a large association with knee flexion peak torque (r = 0.67; 95% CI [0.22, 0.89]; R² = 0.45; large) and knee extension amplitude correlated largely with knee extension total work (r = 0.65; 95% CI [0.16, 0.95]; R² = 0.42; large), but negatively with body mass (r = −0.59; 95% CI [−0.86, −0.09]; R² = 0.35; large).

4. Discussion

The aim of the current study was to characterize the anthropometric profile, motor performance and isokinetic strength of Portuguese high-level female handball players and to examine the associations between these variables. Our main findings can be summarized as follows: (i) performance in countermovement and squat jumps was consistent with values reported for female handball players, with average heights of ~32 and ~30 cm (respectively), while horizontal jump distance approached ~196 cm; [3] (ii) throwing velocities were highest during the jump throw from 9 m (~83 km/h) and showed associations with upper limb strength indicators; (iii) shoulder internal rotation strength was higher than external rotation, with average peak torque values about ~36% greater and average torque ~33% higher; (iv) knee extensor strength exceeded flexor strength by ~97% in both peak and average torque, reflecting a marked dominance of the extensor mechanism; and (v) positive correlations were observed between isokinetic variables and motor performance, such as the association between squat jump height and peak torque of the knee extensors. Taken together, these findings reinforce the biomechanical relevance of upper and lower limb strength in executing explosive technical actions in handball and emphasize the role of muscle profiling in training design.

The anthropometric characteristics observed in the players of this study are consistent with profiles reported for elite-level female handball players, particularly in terms of stature and body composition [6,23]. For example, a study involving high-level Spanish players reported similar average values for height (~171 vs. ~171 cm) and arm span (~171 vs. ~171 cm), although slightly lower values for body mass (~72 vs. ~67 kg) and muscle mass (~31 vs. ~25 kg) when compared to the Portuguese players [6]. In contrast, when compared to the Tunisian national team, Portuguese players showed considerably lower values for body mass (~72 vs. ~86 kg) and height (~171 vs. ~189 cm) [24]. Regarding Italian elite players, Portuguese players demonstrated comparable values for height (~171 vs. ~169 cm), body mass (~72 vs. ~67 kg) and body mass index (~24 vs. ~23 kg·m⁻²) [25]. These findings reflect the morphological adequacy of the Portuguese players to the physical requirements of high-performance handball (suggesting that they possess an anthropometric profile well-aligned with the physiological), technical and tactical demands of the sport.

The players in this study achieved average heights of ~32 cm in the countermovement jump and ~30 cm in the squat jump, which are in line with elite-level standards. Similar averages have been reported in studies with high-level Portuguese players (with ~29 cm and ~28 cm, respectively) [3]. Norwegian players of comparable age also showed performances of ~32 cm and ~30 cm [26]. In contrast, senior players from the Slovenian national team exhibited higher averages of ~39 cm in the countermovement jump and ~34 cm in the squat jump [27]. These findings suggest that Portuguese players possess a competitive level of lower-limb explosive strength. However, their performance still falls short of the values observed in top-tier international squads, which may indicate a need for targeted interventions focused on neuromuscular power development [27]. Regarding horizontal jump performance, the players demonstrated an average distance of ~196 cm, which aligns with international reference values and closely matches results observed in elite Portuguese players (~197 cm) [3]. Conversely, higher averages have been reported in athletes from the Tunisian national team (~249 cm), potentially reflecting differences in morphological traits, training methodology or long-term physical specialization [24]. The Portuguese players demonstrated a solid level of horizontal propulsion aligned with high-level handball demands. However, there is still room for improvement when compared to teams with stronger international performance.

The average throwing velocity was ~76 km/h in the standing throw from 7 m, ~83 km/h in the jump throw from 9 m and ~83 km/h in the standing throw after three steps. Studies involving high-level Portuguese players have shown similar averages (~80 km/h in the 7 m throw and ~83 km/h in the jump throw), indicating a technical and physical performance aligned with the demands of the sport [3]. In contrast, players from the Icelandic national team exhibited slightly lower values, with velocities of ~74 km/h in the standing throw and ~76 km/h in the jump throw [6]. The similarity in velocity between the jump throw and the three-step throw suggests consistent power generation regardless of throwing condition, which may reflect mechanical efficiency and effective kinetic chain coordination, both essential for handball throwing performance [4]. The handgrip strength averaged ~37 kg·f among the players evaluated (with similar values between dominant and non-dominant hands), indicating well-balanced upper limb function. These values are consistent with those reported for elite female handball players, with previous studies describing averages ranging from ~34–38 kg·f in high-level populations [28]. These findings indicate that Portuguese players exhibit adequate upper-limb power and coordination, maintaining consistent release speeds across throwing types and reflecting adaptations to the physical and technical demands of high-level handball.

The isokinetic profiles observed in the present study reflect sport-specific adaptations in elite female handball players. Shoulder internal rotation demonstrated higher values than external rotation across all measured parameters, with peak torque ~36% greater in absolute terms and ~37% greater relative to body mass. Average torque was ~33% higher, total work ~25% greater when normalized to body mass and average power exceeded external rotation by ~19%. These findings are consistent with previous reports in female athletes, where the agonist dominance in internal rotation reflects the repeated demands of overhead throwing [11,15]. While the physiological differences in force production between the internal and external rotators are expected, their force balance should be kept under observation to prevent injury occurring. The current evidence suggests that a force imbalance between internal and rotator muscles may lead to overload pain or injury, as the muscles are unable to properly decelerate the movement [8]. The analyzed female players in this study have an external/internal force ratio identical to that found in an highly successful team in the Portuguese handball league [29]. However, caution is advised when drawing conclusions from this fact, as variables such as sex, training load, and game style and position my contribute to the risk of shoulder injury [8], and also because functional ratios at different test speeds may be more precise. At the knee, extension capacity markedly exceeded flexion, with peak torque nearly double in absolute and relative terms. These results are in line with normative isokinetic profiles reported for elite female handball players, where extensor dominance supports the explosive demands of sprinting, jumping and landing [30,31]. Together, these data reinforce the asymmetric loading patterns inherent to handball and highlight the importance of targeted strengthening of the external rotators and hamstrings to balance performance demands and reduce injury risk, since weakness in the shoulder external rotator muscles significantly increases the probability of shoulder injury in handball athletes [8]. Likewise, hamstring weakness has been linked to a higher risk of hamstring strains (a common sport-related injury), emphasizing the need for specific hamstring strengthening in training programs [32].

The associations observed between isokinetic strength variables and motor performance indicators revealed consistent patterns across both upper and lower limbs, underscoring the critical role of muscular function in the technical performance of handball. Peak torque of the external shoulder rotators showed moderate to strong associations with squat jump performance and throwing velocity (r = 0.540–0.760), while peak torque of the internal rotators was also correlated with the 7 m throw (r = 0.550) and squat jump (r = 0.674). This interrelationship may reflect the importance of trunk and shoulder stabilization in transferring force effectively through the kinetic chain, not only in upper limb actions but also during lower-limb explosive movements. The literature supports that strengthening the shoulder rotators through isokinetic training can enhance throwing velocity and is directly related to performance in similar sport-specific tasks [33,34], reinforcing the rationale for incorporating joint-specific strength interventions into broader athletic development programs.

Regarding the lower limbs, knee extension peak torque was positively correlated with squat jump height (r = 0.650), suggesting that extensor strength is a key determinant of vertical propulsion. Evidence also indicates that this strength is linked to throwing velocity in both standing and jump throw conditions [35], which may reflect the contribution of lower-limb force transmission to kinetic chain efficiency during overhead throwing. A strong negative correlation between body mass and knee extension total work (r = −0.650) suggests that excessive mass may hinder the expression of lower-limb power. In addition, significant interrelationships were found between total work and power variables for both upper and lower limbs, supporting an integrated model of muscle function in sport-specific actions in handball. This integrative perspective aligns with contemporary performance models, which emphasize the coordination and force synergy across multiple joints and segments to achieve efficient technical execution.

Despite offering valuable insights into the anthropometric, motor performance, and isokinetic strength profiles of elite Portuguese female handball players, this study presents several limitations that should be considered. Firstly, the relatively small sample size, although representative of a national-level cohort, considerably limits the generalizability of the results. Caution should be taken when extrapolating these findings to other populations, particularly players from different age groups, competitive levels, or countries. Secondly, the cross-sectional nature of the study restricts the ability to draw causal inferences regarding the relationships between strength characteristics and performance outcomes. Thirdly, isokinetic testing was conducted at a single angular velocity (60°·s⁻¹), which may not reflect the full spectrum of movement velocities required in sport-specific tasks such as throwing and jumping. Finally, the study did not account for potential confounding variables (e.g., players positional roles, training history or accumulated load), which may have influenced the observed associations.

Future research should consider longitudinal or intervention-based designs to explore the causal impact of targeted strength development on handball performance and injury risk. Expanding the sample to include players from different competitive levels, age groups, and positional roles would enhance generalizability. In addition, incorporating multi-velocity isokinetic testing, electromyographic analysis and field-based performance metrics could provide a more comprehensive understanding of the neuromuscular demands in elite handball. Investigating asymmetries, fatigue responses, and their relation to injury mechanisms may also support the development of more individualized and effective training programs.

5. Practical Applications

Given the clear associations between upper- and lower-limb strength and key performance indicators, training programs for elite female handball players should prioritize a systematic approach to the development of different manifestations of strength. This includes maximal strength (e.g., heavy squats, rowing variations, 3–5 sets × 3–6 reps at ≥80% 1RM), explosive strength/power (e.g., loaded jump squats, Olympic lifts, medicine ball throws, 3–4 sets × 3–6 explosive reps at 30–70% 1RM), and muscular endurance/stability (e.g., banded external rotations, Nordic hamstring exercise, 2–3 sets × 12–15 reps). These examples are provided as practical suggestions and should be adapted to the individual athlete’s physical condition, performance level, and injury history. Proper periodization and integration of these components are essential to maximize performance adaptations while reducing fatigue and injury risk.

Isokinetic assessment data should be used to guide individualized training loads (particularly in athletes with low power-to-body mass ratios or asymmetries between limbs). Monitoring these parameters can inform targeted interventions and injury prevention strategies. In addition, considering the neuromuscular demands of the sport, complex training methods (e.g., combining heavy strength with plyometric or ballistic work) are recommended to improve rate of force development and technical execution under fatigue. Alongside isokinetic data, anthropometric characteristics such as height, arm span, and body composition also play a decisive role in performance. Taller players with greater reach may achieve higher throwing efficiency, while optimal body composition supports repeated jump performance and resistance to fatigue. Systematic monitoring of these variables can help guide individualized conditioning programs and inform nutritional and recovery strategies in elite handball. Altogether, these findings emphasize the value of integrated physical profiling not only for optimizing performance but also for designing individualized programs that minimize functional limitations in elite female handball players.

6. Conclusions

This study provides an integrated profile of anthropometric, motor, and isokinetic characteristics in elite Portuguese female handball players, highlighting clear associations between upper- and lower-limb strength and sport-specific performance. Rotational shoulder strength and knee extensor capacity were especially relevant for throwing and jumping tasks. The data reinforce the value of isokinetic testing in identifying functional asymmetries and guiding individualized training. Despite limitations, the findings contribute important reference values for performance monitoring and injury prevention in high-level female handball.