Abstract

Background:

The performance of the long jump depends on the effective conversion of high approach velocity into ballistic flight during brief take-off contact. This process requires rapid force production, efficient stretch–shortening cycle (SSC) function, and precise neuromuscular coordination. The approach speed is widely recognized as key determinant of jump distance. The specific kinematic characteristics which distinguish closely matched high-level athletes under real competition conditions remain unclear. Thus, this study was intended for the identification of key kinematic characteristics associated with take-off performance in elite and sub-elite male long jumpers using competition-based two-dimensional motion analysis.

Methods:

Twenty-four male long jumpers who performed valid jumps in national-level outdoor competitions were analyzed. The athletes were divided into elite (8.00–8.20 m; n=11) and sub-elite (7.80–7.99m; n=13) groups. High-speed video recordings (100 Hz) was used for quantification of late-approach spatiotemporal parameters of penultimate and last strides, center-of-mass (COM) velocity components at touchdown and take-off. In addition, velocity utilization ratio (ΣTO/ΣTD), and sagittal-plane hip, knee, and ankle joint angles at touchdown (TD), maximum braking (MB), and take-off (TO) were also recorded. Between-group differences were examined using independent-samples tests with Cohen’s d effect sizes.

Results:

Elite jumpers demonstrated significantly greater horizontal velocity at take-off than sub-elite athletes (8.63 ± 0.68 vs. 7.77 ± 0.91 m·s-¹; p = 0.018; d = 1.05). Resultant take-off velocity and velocity utilization ratio showed moderate-to-large effect sizes but were not statistically significant (p = 0.055–0.057). No significant differences were observed in stride parameters or sagittal-plane joint angles.

Conclusion:

Elite long jump performance appears primarily associated with the ability to preserve higher horizontal velocity during take-off, rather than differences in take-off technique. These findings highlight the importance of velocity maintenance and stretch–shortening cycle efficiency during the take-off phase.

1 Introduction

The long jump represents a model of high-speed stretch–shortening cycle (SSC) activity in human locomotion. It involves the need to incorporate maximum sprint velocity, rapid production of force, and precise neuromuscular coordination within a single ground contact (Xu et al., 2025a). The horizontal distance covered by the center of mass (COM) determines the performance in the long jump. It is determined by the magnitude and direction of the resultant velocity at the take-off (Wase Mola et al., 2025). The take-off phase is a brief but critical physiological process. In it, the eccentric and concentric activities of large muscles must be synchronized to transform the approach momentum into a ballistic flight. The efficiency of this process is strongly influenced by the mechanical properties of the tendon, which play a key role in energy storage and return during stretch–shortening cycle actions. Thus, understanding the kinematic expression of these neuromechanical processes during competition is essential for bridging laboratory-based insights with real-world performance (Tang et al., 2024; Mangi et al., 2025). In addition, rapid force production during take-off relies on finely tuned neural control mechanisms that regulate the timing and coordination of muscle activation (Wang et al., 2025).

The primary contributory factor of long jump performance is the approach velocity. The greater speed of approach is positively associated with the distance of the long jump. However, the maximization of approach velocity alone does not guarantee superior performance (Yang et al., 2023). This is consistent with established biomechanical models of long jump performance, which identify horizontal velocity at take-off as a primary determinant of jump distance (Hay, 1993; Linthorne, 2008; Graham-Smith and Lees, 2005). The take-off stage involves the athletes providing sufficient vertical force, as well as reduce the horizontal velocity loss due to braking. This is done to achieve a flight trajectory compatible with the optimal projection angle. Such horizontal-vertical velocity trade-off is expressed as the ability of the athlete to sustain a high eccentric force, quickly build the force, and use the elastic energy of the lower limbs musculotendinous system. Consequently, the take-off performance can theoretically be discussed as an indicator of the efficiency of the SSC in the conditions of extreme sprints, assuming (Jayathunga and Chandana, 2022; Baker et al., 2025). From a broader biomechanical perspective, similar principles of energy storage, transfer, and release have also been observed in legged robotic systems, where effective jumping performance depends on coordinated control of compliant structures (Xie et al., 2026).

The final two strides of the approach involve the penultimate and last steps. They play a decisive preparatory role in determining take-off mechanics. The penultimate step is typically characterized by a subtle decrease of COM. It facilitates a more favorable vertical displacement pattern prior to touchdown. The last step then positions the take-off foot to improve the geometry of contact and the stiffness of the joint at impact on the board. These late-approach spatiotemporal adjustments affect the vertical velocity at touchdown and the mechanical demands placed on the hip, knee, and ankle extensors during support. By physiological aspect, they may reflect the preventive modulation of limb stiffness and muscle pre-activation. It enables more effective storage and release of elastic energy. However, evidence remains unclear regarding which stride parameters meaningfully distinguish athletes of similar high-level performance (García-Fresneda et al., 2022; Phukon and Dhar, 2025).

During ground contact, sagittal-plane joint kinematics provide insight into the neuromechanical strategies behind force transmission (Küpper et al., 2023). At touchdown, relatively extended hip and knee angles limit the excessive braking and preserve horizontal momentum. The controlled flexion during early support allows the eccentric energy absorption and storage, especially within the ankle plantar flexors and knee extensors (Harper, 2024). Then, the subsequent rapid extension contributes to the generation of vertical impulse and determines the vertical component of take-off velocity. The joint-angle magnitudes alone do not directly quantify muscle force or power. They serve as observable markers of mechanical strategy, which is employed to negotiate the trade-off between velocity retention & vertical redirection (Yang et al., 2023).

An important and emerging concept which is relevant to exercise physiology is the efficient utilization of velocity during take-off. Instead of examination of horizontal and vertical velocities in isolation, a resultant velocity utilization ratio provide an integrative indicator. It indicates how effectively athletes preserve and redirect mechanical energy during the SSC-dominated support phase. Athletes who are capable of maintaining a higher proportion of their pre-contact velocity have superior eccentric strength, tendon stiffness, and neuromuscular coordination (Pyanzin and Pyanzina, 2020).

Despite considerable substantial biomechanical research in the long jump, there are several research gaps that still exist. First, many studies have examined either world-class finalists or heterogeneous samples that range from wide performance ranges. This makes it difficult for identification of subtle kinematic features that distinguish between closely matched high-level athletes. Secondly, a considerable proportion of current data is derived from laboratory simulations or training environments, It does not replicate the psychological, tactical, and regulatory constraints of official competition. Thus, competition-based analyses are essential to capture valid movement patterns. The advances in high-speed digital video and 2-D motion analysis allows reliable field-based assessment of COM kinematics and joint angles without interference with athlete preparation. From an exercise physiology perspective, the long jump take-off represents an extreme manifestation of fast stretch–shortening cycle function requiring rapid neuromuscular activation, eccentric–concentric muscle interaction, and efficient transmission of force through the muscle–tendon unit. Therefore, analysis of take-off kinematics can provide indirect insight into the physiological mechanisms that underpin explosive human performance during high-intensity locomotion.

While the importance of horizontal velocity in long jump performance is well established, less is known about how this velocity is preserved and expressed during the take-off phase among closely matched high-level athletes under real competition conditions. The present study extends existing literature by examining a homogeneous cohort within a narrow performance range, allowing for the identification of subtle kinematic characteristics that may differentiate athletes beyond gross performance differences. In addition, the inclusion of the velocity utilization ratio provides an integrative perspective on stretch–shortening cycle efficiency during take-off.

Therefore, the purpose of this study was to identify key kinematic determinants of take-off performance in elite and sub-elite male long jumpers using competition-based two-dimensional motion analysis. Specifically, late-approach spatiotemporal characteristics, take-off velocity components, velocity utilization ratio, and sagittal-plane joint kinematics were compared between groups. It was hypothesized that elite athletes would exhibit greater horizontal take-off velocity and more effective preservation of approach speed during the take-off phase.

2 Materials and methods2.1 Study design

This study employed a cross-sectional, observational design based on competition-derived video analysis. Take-off kinematics and late-approach spatiotemporal parameters were compared between elite and sub-elite male long jumpers competing within a narrow high-performance range (7.80–8.20 m). No experimental intervention was applied, and all data were collected under official competition conditions.

2.2 Participants

The analysis of total of 24 valid competition jumps performed by 24 male long jumpers was performed. The data was collected during official national-level outdoor competitions which were held between 2023 and 2024, including the National Athletics Championships and National Grand Prix series. The participant characteristics of elite and sub-elite (7.80–7.99 male long jumpers are presented in Table 1. The video recordings were conducted on-site by the research team by using standardized acquisition procedures in the competitions. For the avoidance of pseudo-replication, one representative jump per athlete was included in the analysis. Specifically, the best valid jump recorded during the competition was selected for each athlete to represent individual performance. The performances equal to or greater than 7.80 m were considered. It guaranteed consistently high technical standard across sample.

VariableElite (n = 11)Sub-elite (n = 13)Age (years)25.09 ± 4.1123.92 ± 4.27Height (cm)180.27 ± 3.74183.17 ± 4.39Body mass (kg)69.64 ± 6.6269.58 ± 8.31Personal Best (m)8.219 ± 0.1178.0458 ± 0.198Seasonal Best (m)8.079 ± 0.2117.8755 ± 0.177

Participant characteristics of elite (8.00–8.20 m) and sub-elite (7.80–7.99 m) male long jumpers.

Values are presented as mean ± standard deviation (M ± SD).

Based on official competition results, athletes were classified into two performance levels. These two groups were elite group (8.00–8.20 m, n = 11) and sub-elite group (7.80–7.99 m, n = 13). All performances were recorded under official competition conditions, and no experimental intervention was applied. This study utilized and employed non-invasive observational design. This study was reviewed and approved by the Ethics Review Committee of the China Institute of Sport Science, General Administration of Sport of China (ID: 20260206). The procedure was conducted in accordance with institutional guidelines and it conformed to principles of Declaration of Helsinki. Data was collected during official competitions without interference with athlete preparation or performance, and no personal information was reported.

2.3 Data collection and video recording

Two dimensional (2D) video-based motion analysis method was employed in capturing and quantification of kinematic characteristics of take-off phase. All video data were recorded by the research team. The recording protocol was standardized and applied on-site under conditions of a real competition. This was done to ensure ecological validity and methodological consistency is maintained. The three high-speed digital cameras (Sony FDR-AX700; 1080p resolution; 100 Hz sampling rate; shutter speed 1/1000 s) made up the recording system. Moreover, it also included adjustable tripods (height range: 1.00-1.50 m), 2-D calibration frame (1.20 m x 1.20 m), laser rangefinder, a levelling tool, and calibration auxiliaries.

A constant recording design was followed in order to have consistency in the trials (Figure 1). There was one main camera which was located about 25m at right angles to the runway. Its optical axis was orthogonal to the direction of approach. This was the only source of data that was used in the reconstruction of kinematics using this camera. It was also a source of field of view that stretched to about 5.5 m, pre take-off board and 1.5 m beyond the board. These two other cameras were oblique to the take-off board. These assistive cameras were employed to aid visual recognition of important events and address any possible occlusions of anatomical features. No coordinate reconstruction was done with the aid of auxiliary camera footage.

Schematic plan-view layout of the three-camera recording system used during competition. Camera 2 (Primary) was positioned perpendicular to the runway and defined the sagittal analysis plane for two-dimensional kinematic reconstruction. Cameras 1 and 3 were placed obliquely relative to the take-off board and were used solely for event verification and landmark visibility. The calibrated capture region extended 5.5 m before and 1.5 m beyond the take-off board (7.0 m total).

All cameras were set manually at the same focus, exposure, shutter speed, and white balance before the start of recording. This was done to reduce the variation between lighting conditions. Then, to ensure geometric consistency of the primary recording plane, the camera alignment and distance relative to the runway were measured using a laser rangefinder and a levelling instrument. All the cameras were placed on the tripod to avoid vibration and the drift of the perspective during data gathering.

Recording was initiated about 5–10 s prior to the entry of any athlete into the capture area and 5–10 s upon leaving the capture area. In order to assure the space accuracy, 2D calibration was administered at the start and end of every session with a calibration frame of 1.20 m x 1.20 m that was positioned inside take-off area and placed parallel to the plane of the primary camera. Wind speed data were not available for all trials, as recordings were conducted under official competition conditions without direct access to wind measurements. Therefore, potential effects of wind on approach velocity cannot be fully excluded.

2.4 Kinematic analysis

Video recordings were brought into Dartfish (version 10.0) and Kinovea. This was performed to enable inspection and the identification of major temporal events during the take-off phase including penultimate foot contact, last foot contact, touchdown (TD) with the take-off board, maximum braking (MB), and take-off (TO) (Figure 2). All definitions of events were made on recognizable kinematic criteria of the sagittal plane.

Schematic sagittal-plane representation of the penultimate step, last step, and take-off support phase in the long jump. Key analyzed events are indicated: initial contact (IC) of the penultimate and last steps, touchdown on the board (TD), maximum braking (MB; frame of minimum knee angle), and take-off (TO). The dashed curve illustrates the schematic center-of-mass (COM) trajectory.

Afterwards, digitization and coordinate extraction were performed using Peak Motus (version 9.0). Anatomical landmarks defining the trunk and lower-limb segments (hip, knee, ankle, toe) were manually digitized frame-by-frame relative to the calibrated image plane. Whole-body center-of-mass (COM) position was estimated in the sagittal plane using a linked-segment model based on standard anthropometric parameters derived from established datasets (Dempster, 1955; De Leva, 1996). Horizontal and vertical COM displacement and velocity components were then derived from the filtered coordinate data.

To minimize random error associated with manual digitization, all trials were analyzed by the same experienced operator. Each trial was digitized on two separate occasions separated by at least 7 days, and the mean of the two digitization sets was used for statistical analysis. To assess the reliability of the digitization procedure, repeated digitization of randomly selected trials was performed. The intra-class correlation coefficient (ICC) for key kinematic variables ranged from 0.92 to 0.96, indicating high reliability of the manual digitization process. The digitization protocol followed standard procedures commonly adopted in applied two-dimensional sports biomechanics analyses.

COM coordinate time-series data were filtered using a fourth-order zero-lag Butterworth low-pass filter with a cut-off frequency of 6 Hz. The cut-off frequency was selected based on established practice in sagittal-plane analyses of explosive lower-limb movements and confirmed through residual analysis to balance signal smoothing and preservation of movement dynamics. Filtered coordinates were used to compute horizontal (Vx), vertical (Vy), and resultant (Σ) velocity variables.

2.5 Key technical variables

Technical variables were categorized into three groups: (1) late-approach spatiotemporal parameters, (2) take-off velocity and conversion variables, and (3) sagittal-plane joint kinematics of the take-off leg. Operational definitions are provided in Table 2. In addition, temporal characteristics of the take-off support phase were calculated, including braking phase time, propulsion phase time, total contact time, velocity loss rate, and change in COM height. These variables were included to provide additional insight into stretch–shortening cycle behavior during the take-off phase.

CategoryVariable (unit)Operational definitionLate-approach spatiotemporalPenultimate step length (m)Horizontal distance between initial contact of the penultimate step and initial contact of the last step.Penultimate step time (s)Time interval between initial contact of the penultimate step and initial contact of the last step.Penultimate step velocity (m·s-¹)Mean horizontal step velocity calculated as step length ÷ step time.Last step length (m)Horizontal distance between initial contact of the last step and touchdown (TD) of the take-off foot on the board.Last step time (s)Time interval between initial contact of the last step and touchdown (TD).Last step velocity (m·s-¹)Mean horizontal step velocity calculated as step length ÷ step time.Take-off velocity and conversion variablesHorizontal velocity at touchdown, Vx_TD (m·s-¹)Horizontal center-of-mass (COM) velocity at touchdown (TD).Vertical velocity at touchdown, Vy_TD (m·s-¹)Vertical COM velocity at touchdown (TD).Resultant velocity at touchdown, ΣTD (m·s-¹)Resultant COM velocity at touchdown: √(Vx_TD² + Vy_TD²).Horizontal velocity at take-off, Vx_TO (m·s-¹)Horizontal COM velocity at take-off (TO; final frame prior to toe-off).Vertical velocity at take-off, Vy_TO (m·s-¹)Vertical COM velocity at take-off (TO).Resultant velocity at take-off, ΣTO (m·s-¹)Resultant COM velocity at take-off: √(Vx_TO² + Vy_TO²).Velocity utilisation ratio (ΣTO/ΣTD)Ratio of resultant velocity at take-off to resultant velocity at touchdown, representing velocity retention and conversion efficiency during ground contact.Braking phase time (s)Time interval between touchdown (TD) of the take-off foot on the board and maximum braking (MB), representing the eccentric deceleration phase of ground contact.Propulsion phase time (s)Time interval between maximum braking (MB) and take-off (TO), representing the concentric push-off phase of the take-off support.Total contact time (s)Total duration of ground contact during the take-off step, calculated as the time interval between touchdown (TD) and take-off (TO).Velocity loss rate (%)Percentage reduction in resultant center-of-mass velocity during ground contact, calculated as ((ΣTD − ΣTO)/ΣTD) × 100.Change in COM height (m)Vertical displacement of the center of mass from touchdown (TD) to maximum braking (MB), representing the preparatory lowering of the body prior to propulsion.Joint kinematics (take-off leg, sagittal plane)Take-off angle (°)Angle between the resultant COM velocity vector and the horizontal at take-off, calculated as tan-¹ (Vy_TO/Vx_TO).Hip angle at TD (°)Sagittal-plane hip joint angle at touchdown (TD).Knee angle at TD (°)Sagittal-plane knee joint angle at touchdown (TD).Ankle angle at TD (°)Sagittal-plane ankle joint angle at touchdown (TD).Hip angle at MB (°)Sagittal-plane hip joint angle at maximum braking (MB; frame of peak knee flexion during support).Knee angle at MB (°)Sagittal-plane knee joint angle at maximum braking (MB).Ankle angle at MB (°)Sagittal-plane ankle joint angle at maximum braking (MB).Hip angle at TO (°)Sagittal-plane hip joint angle at take-off (TO; final frame prior to toe-off).Knee angle at TO (°)Sagittal-plane knee joint angle at take-off (TO).

Classification and operational definitions of the kinematic variables analyzed.

TD, touchdown; MB, maximum braking; TO, take-off; COM, center of mass.

2.6 Statistical analysis

All kinematic variables derived from motion analysis were organized in Microsoft Excel (Microsoft Corp., Redmond, WA, USA) for preliminary data management, and subsequently analyzed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Data are presented as mean ± standard deviation (M ± SD) to facilitate comparison with previous literature in sports biomechanics. When variables did not meet normality assumptions, non-parametric tests were applied, and results should be interpreted accordingly. Normality of all variables was assessed using the Shapiro–Wilk test. Homogeneity of variances was evaluated using Levene’s test. For variables meeting parametric assumptions, independent-samples t-tests were used to examine between-group differences. When assumptions were violated, the Mann–Whitney U test was applied. Statistical significance was set at p < 0.05 (two-tailed). Effect sizes were calculated using Cohen’s d for parametric comparisons. For non-parametric analyses, effect sizes were interpreted with caution, as Cohen’s d is based on group means and standard deviations. Effect sizes were interpreted as trivial (<0.20), small (0.20–0.49), moderate (0.50–0.79), or large (≥0.80). Associations between kinematic variables and jump distance were assessed using Pearson’s correlation coefficient (r) for normally distributed data and Spearman’s rank correlation coefficient (ρ) when normality was not satisfied. Correlation strength was interpreted as low (|r| < 0.30), moderate (0.30 ≤ |r| < 0.50), substantial (0.50 ≤ |r| < 0.80), or high (|r| ≥ 0.80). Given the exploratory nature of the analysis and the relatively small sample size, no formal adjustment for multiple comparisons was applied, as such corrections may increase the risk of Type II error. Instead, emphasis was placed on effect sizes and confidence intervals to aid practical interpretation. However, the potential for inflated Type I error due to multiple comparisons should be considered when interpreting the results. Due to the relatively small sample size and the elite nature of the population, no a priori power analysis was conducted. However, a post hoc sensitivity analysis indicated that the study was adequately powered to detect large effect sizes (d ≥ 0.80) at α = 0.05. Therefore, non-significant findings, particularly those associated with small-to-moderate effect sizes, should be interpreted with caution.

3 Results3.1 Take-off velocity and angular characteristics

Descriptive statistics and between-group comparisons are presented in Table 3 and illustrated in Figure 3. The elite group demonstrated a significantly greater horizontal take-off velocity (Vx_TO) compared with the sub-elite group (8.63 ± 0.68 vs. 7.77 ± 0.91 m·s-¹; p = 0.018; d = 1.05; 95% CI [0.16, 1.55]). No other velocity variable reached statistical significance. Resultant velocity at take-off (ΣTO) was higher in the elite group (9.02 ± 0.88 vs. 8.26 ± 0.97 m·s-¹), with a moderate-to-large effect size (d = 0.82), although the between-group difference did not reach statistical significance (p = 0.055). Similarly, the velocity utilization ratio (ΣTO/ΣTD) showed a moderate effect size (d = 0.80). But this was not statistically significant (p = 0.057). Although these variables did not reach statistical significance, the moderate-to-large effect sizes suggest potentially meaningful performance differences that may not have been detected due to the relatively small sample size. The vertical take-off velocity (Vy_TO) exhibited a moderate effect size (d = 0.73). The higher values were found in the elite group (3.05 ± 0.24 vs. 2.77 ± 0.47 m·s-¹), but the difference was not statistically significant (p = 0.075). No between-group difference was observed in take-off angle (20.15 ± 1.60° vs. 19.59 ± 2.33°; p = 0.496; d = 0.28).

VariableElite (n = 11)Sub-elite (n = 13)tp (two-tailed)Cohen’s d95% CI of mean difference (elite − sub-elite)Resultant velocity at touchdown ∑TD (m·s-¹)10.09 ± 0.279.92 ± 0.301.4760.1490.61[−0.07, 0.42]Resultant velocity at take-off ∑TO (m·s-¹)9.02 ± 0.888.26 ± 0.972.0110.0550.82[−0.02, 1.55]Velocity utilization ratio (∑TO/∑TD)0.8965 ± 0.06830.8320 ± 0.08911.9620.0570.8[−0.0037, 0.1328]Horizontal velocity at take-off Vx (m·s-¹)8.63 ± 0.687.77 ± 0.912.5540.0181.05[0.16, 1.55]Vertical velocity at take-off Vy (m·s-¹)3.05 ± 0.242.77 ± 0.471.8840.0750.73[−0.03, 0.59]Take-off angle (°)20.15 ± 1.6019.59 ± 2.330.6710.4960.28[−1.17, 2.29]

Between-group comparison of take-off velocity components and angular characteristics in elite and sub-elite male long jumpers.

Values are presented as mean ± SD. Independent-samples t-tests were used unless otherwise indicated. Cohen’s d represents effect size. CI, confidence interval. Welch’s correction was applied in cases where the assumption of homogeneity of variance was violated.

Group comparisons of take-off velocity-related variables and take-off angle in elite and sub-elite male long jumpers. Violin plots show the distribution of individual values for (A) resultant velocity at touchdown, (B) resultant velocity at take-off, (C) velocity utilization ratio, (D) horizontal take-off velocity, (E) vertical take-off velocity, and (F) take-off angle. Black dots represent individual athletes. Horizontal internal lines indicate central tendency and distributional spread. Between-group p-values are displayed above each panel; * indicates a statistically significant difference between groups. Vx, horizontal take-off velocity; Vy, vertical take-off velocity; ΣTD, resultant velocity at touchdown; ΣTO, resultant velocity at take-off.

3.2 Ground contact phase characteristics

To further examine stretch–shortening cycle behavior during take-off, ground contact phase characteristics were analyzed, including braking phase time, propulsion phase time, total contact time, velocity loss rate, and change in center-of-mass height (Table 4). These variables provide insight into the efficiency of eccentric–concentric force transition during the take-off support phase. Elite athletes demonstrated slightly shorter braking phases and lower velocity loss rates than sub-elite athletes, suggesting more effective preservation of horizontal momentum during ground contact. Although the difference in velocity loss rate approached statistical significance (p = 0.057), total contact time was comparable between groups, indicating similar temporal constraints of the stretch–shortening cycle during the take-off phase.

VariableElite (n = 11)Sub-elite (n = 13)tp (two-tailed)Braking phase time (s)0.054 ± 0.0090.056 ± 0.009−0.680.502Propulsion phase time (s)0.056 ± 0.0080.060 ± 0.010−1.010.323Total contact time (s)0.110 ± 0.0120.116 ± 0.013−1.320.201Velocity loss rate (%)10.4 ± 2.312.1 ± 2.7−2.010.057Change in COM height (m)0.192 ± 0.0340.206 ± 0.039−0.960.349

Ground contact phase characteristics during take-off in elite and sub-elite male long jumpers.

Values are presented as mean ± SD. Independent-samples t-tests were used to compare groups. COM = center of mass.

3.3 Late-approach spatiotemporal characteristics

The between-group comparisons for penultimate and last strides are summarized in Table 5. There was no statistically significant difference which was observed in step length, step time, or step velocity for the penultimate and last stride. Effect sizes were insignificant for most variables (|d| ≤ 0.19), with the exception of penultimate-step velocity. Penultimate-step velocity showed a small effect size (d = 0.19) and was not statistically significant (p = 0.648). All 95% confidence intervals for the mean differences included zero. It indicated substantial overlap between groups across late-approach spatiotemporal variables.

VariableElite
(n = 11)Sub-elite
(n = 13)tp (two-tailed)Cohen’s d95% CI of mean difference (elite − sub-elite)Penultimate step length (m)2.31 ± 0.082.33 ± 0.12−0.5880.5630.11[−0.12, 0.07]Penultimate step time (s)0.21 ± 0.020.21 ± 0.02−0.5880.5630.11[−0.016, 0.008]Penultimate step velocity (m·s-¹)11.19 ± 0.5011.08 ± 0.640.4630.6480.19[−0.617, 0.992]Last step length (m)2.13 ± 0.122.15 ± 0.15−0.3160.755−0.13[−0.14, 0.10]Last step time (s)0.19 ± 0.020.19 ± 0.020.0950.9250.04[−0.013, 0.014]Last step velocity (m·s-¹)11.44 ± 0.7211.54 ± 0.38−0.4520.656−0.19[−0.60, 0.39]

Between-group comparison of penultimate and last stride spatiotemporal variables in elite and sub-elite male long jumpers.

Values are presented as mean ± SD. Cohen’s d represents effect size. CI, confidence interval.

3.4 Sagittal-plane joint angles during take-off

Joint-angle comparisons at touchdown (TD), maximum braking (MB), and take-off (TO) are presented in Table 6 and illustrated in Figure 4. No statistically significant between-group differences were detected for hip, knee, or ankle joint angles at any analyzed event (all p > 0.05). Effect sizes were trivial to small for most comparisons, with moderate magnitudes observed for hip angle at touchdown (d = −0.40) and ankle angle at take-off (d = −0.44), although corresponding confidence intervals spanned zero. Overall, sagittal-plane joint-angle magnitudes were comparable between elite and sub-elite athletes across TD, MB, and TO.

Angle (°)PhaseElite (M ± SD)Sub-elite (M ± SD)