Effect of Consecutive Jumping Trials on Metatarsophalangeal, Ankle, Knee and Hip Biomechanics During Take‐Off and Landing: A Replication and Extension Study Authors: Youri Duchene, Nicolas C Hanen, Jennifer Murphy, Léo Banet, Gérome C Gauchard, Guillaume Mornieux Published: 2026 ABSTRACT ABSTRACTReplication is a fundamental aspect of scientific research, but few replications have been conducted within sports biomechanics yet. The aim of this study was to closely replicate Lam et al. (2021) study, investigating the differences of single and consecutive jump landing trials on ground reaction forces, metatarsophalangeal joint, ankle joint and knee joint kinematics and kinetics. We extended the analysis by including the hip joint biomechanics. Forty basketball players performed single and consecutive countermovement jumps (CMJ) tasks. Synchronised motion capture system and force plates were used to compute biomechanical variables during take‐off and landing. Paired t‐tests were performed to examine any significant effect of the task. The compatibility of the replication and original effect size estimates was evaluated with z‐tests. The consecutive jumping task led to lower jump height, takeoff velocity and vertical ground reaction forces at landing for both studies. Unlike the original study, metatarsophalangeal joint, ankle joint, and knee joint velocities and powers were greater in single countermovement jump condition compared to consecutive condition. Additionally, the hip joint displayed larger velocities and power during take‐off in the single CMJ condition compared to consecutive CMJ condition. The present study couldn't replicate most of the original findings, as the effect of consecutive jump on most of the joint kinematics and kinetics results was not consistent between the two studies. We recommend authors to describe the level of practice of their population and to adopt Open Science principles to build higher confidence in the sports sciences results.Keywords: basketball, countermovement jumps, inverse dynamics, lower limb, replicability Introduction Replication involves retesting a claim using the same analyses with new data, where results that support the original study findings should increase confidence in the claim (Nosek and Errington 2020). Replication studies are rare in sport and exercise science (Mesquida et al. 2022), despite calls for an increased number of them (Caldwell et al. 2020; Halperin et al. 2018) and concerns about the replicability of sports and exercise science research (Borg et al. 2020; Caldwell et al. 2020; Mesquida et al. 2022). In particular, there is a high positive findings results rate (81%) (Twomey et al. 2021) meaning that statistically significant findings are mostly published in this research field. Yet, given the tentative evidence for publication bias in the field (Mesquida et al. 2022; Borg et al. 2023) and typical small sample sizes (Abt et al. 2020), it is clear why many sports and exercise science researchers subjectively believe that replication could be an issue (Murphy et al. 2023). The Sports Science Replication Centre was established (see https://ssreplicationcentre.com/) to systematically evaluate the replicability of sport and exercise science research via a worldwide, collaborative large‐scale project (Murphy et al. 2025). A formalized protocol was published to transparently describe how studies are selected for replication, considering statistical, theoretical and methodological factors (Murphy, Mesquida, Caldwell, et al. 2023; Murphy et al. 2025). Consequently, we were allocated a study titled “Effect of consecutive jumping trials on metatarsophalangeal, ankle, and knee biomechanics during take‐off and landing”, published in the European Journal of Sports Sciences (Lam et al. 2021). This study investigated the effect of consecutive jumps versus single jump on lower limb biomechanics and performance in basketball. Jumping is one major physical component in basketball, either in defensive (i.e., defensive rebounds, block shots) or offensive (i.e., offensive rebounds, jump shots, lay‐ups) situations. In competitive settings, basketball players perform around 45 jumps including both consecutive jumps (e.g., multiple rebounds) and single jumps (e.g., jump shot) (Ben Abdelkrim et al. 2007; McInnes et al. 1995). It is also worth noting that jumping tasks have been widely performed to screen for performance (Bishop et al. 2017; Johnston et al. 2015; Maloney et al. 2017) and injury risk (Hewett et al. 2016; Pontillo et al. 2021). Regarding performance, most of the screening tasks are performed with single jumps, but consecutive jumps have shown to induce lower jump height, ground reaction forces, and a modification in lower limb kinematics (Cormack et al. 2008) and might better represent in‐game situations. Then, landing tasks are mostly used to screen for potential injuries, but recent recommendations lead to have sport‐specific landing tasks (Beardt et al. 2018). Moreover, a consecutive jumping task combined with a single countermovement jump (CMJ) has been related to ACL injury (Collings et al. 2022). Therefore, the lower limb's joint mechanical differences between single and consecutive jumps could further help to understand determinants of jumping performance and to select better screening tasks for injury prevention purposes. Among these joint mechanical aspects, the metatarsophalangeal joint (MTPJ) stiffness was found to impact consecutive jumps performance but no single jumps (Jia et al. 2022). Therefore, this joint should be of particular interest when studying jumping biomechanics. Hence, the objective of the original study (Lam et al. 2021) was to investigate the differences in single and consecutive jump landing trials in terms of ground reaction forces, MTP, ankle and knee kinematics and kinetics during jumping take‐off and landing phases. Our aim was to closely replicate this study and to extend their aim with the kinematic and kinetic analysis of the hip joint, given its major contribution to jump height (Johnston et al. 2015). Materials and Methods The method of the present study was designed to be as similar as possible to the original method, following the framework of Brandt et al. (2014). This study was pre‐registered on the Open Science Framework. A within‐group study design was used to investigate the differences in single and consecutive jump landing trials. The experimental design remained the same as the original, where all participants performed six trials of countermovement jumps in both single and consecutive jump sessions. The original authors were contacted regarding the deidentified raw data and to provide clarity on some methodological queries (i.e., the markerset and the definition of joint moments), but they were not responsive. Original effect size estimates can sometimes be inflated due to publication bias and low statistical power (Anderson and Maxwell 2017), particularly in sports and exercise science where sample sizes are small (Abt et al. 2020; Knudson 2017). Therefore, a priori statistical power calculations based on these can result in statistical power lower than intended. As per the study selection protocol (Murphy, Mesquida, Caldwell, et al. 2023), we used the lower bound 95% confidence interval of the original effect size of the key variable (i.e., peak GRF at landing, Cohen's d z = 1.21 [0.59, 1.82]) at 95% power, α = 0.05, for a paired t‐test which resulted in a required sample size of 40 participants. See our supplementary material for full calculations and sample size justification (https://doi.org/10.17605/OSF.IO/BNVYK). Forty male basketball athletes (Table 1) were recruited from French university basketball teams and clubs. All participants practiced basketball at least twice a week (3 trainings per week in average), competed for the university and/or their club and were free from any lower limb injury in the 6 months before their participation. The average number of years of practice was 11 ± 5 years. The participants were playing at competitive level and could be classified as Tier 2 (Trained/Developmental) for 30 of them and Tier 3 (Highly‐Trained/National level) for the other 10 participants (McKay et al. 2022). Lower limb dominance was determined by asking the athletes to kick a ball placed in front of them, with the dominant leg defined as the one used to kick the ball. The study was approved by the ethics committee Nord Ouest III (approval reference 2023‐A00362‐43) and was conformed to requirements stipulated in the Declaration of Helsinki. This study was pre‐registered on the Open Science Framework (https://doi.org/10.17605/OSF.IO/BNVYK). All participants were informed about the procedures, purpose and possible risks associated with the experimental set‐up and gave their written consent prior to testing. TABLE 1Descriptive characteristics of participants in the original and replication studies.Original studyReplication studyComparisonEffect size n = 18 n = 40 p Cohen's d Age (years)20.9 (1.0)21.5 (3.1)0.6330.24Height (m)1.80 (0.04)1.83 (0.08) 0.010 0.53 Body mass (kg)69.2 (7.7)75.3 (10.0) < 0.001 0.68 Note: Data are presented as mean (standard deviation). Bold values indicate statistical significance (p < 0.05). Descriptive characteristics of participants in the original and replication studies. Note: Data are presented as mean (standard deviation). Bold values indicate statistical significance (p < 0.05). Procedure All athletes performed the exact same protocol as in Lam et al. (2021). After a 10‐min warm‐up and task familiarization, participants performed two jumping conditions, consisting of a single CMJ and six consecutive CMJs. The order of the conditions was randomized. Between conditions, a 10‐min resting period was administered. All athletes repeated each condition 6 times (6 single CMJ, 6 sets of consecutive CMJ) with 1.5 min of rest between each trial. Even though the hand position was not mentioned, participants were asked to place them on the hips, as mentioned in the Cormack et al. (2008) reference. Countermovement depth was self‐selected by the subject. For the consecutive jump condition, the participants were told to perform countermovement jumps, by initiating the next jump immediately after landing. All participants wore standard shoes (Gel‐rocket 10, Asics), which differed from those used in the original study (Li Ning Wade Cloud cushion, Beijing, China), as the latter were not available in France. Although Gel‐Rocket shoes are also designed for indoor team sports, their technical characteristics may slightly differ from those of the original shoes (https://doi.org/10.17605/OSF.IO/BNVYK). A total of 22 reflective markers were placed following Lam et al.’s markerset (2021). Due to unclear information related to the forefoot markerset, we replicated the one described in their previous paper (Lam et al. 2018) and enabling the metatarsophalangeal joint modeling. To do so, the second metatarsal head marker was replaced by a marker on the superior aspect of the hallux. Participants performed the jumps on a force plate (1000 Hz, AMTI, Watertown, USA), synchronized with a 14‐camera motion capture system (200 Hz, Arqus 12MP, Qualisys, Göteborg, Sweden) capturing the 3D reflective markers position during all phases of the jumping tasks. Any deviations from the original study's methodology are transparently reported on the project page (https://doi.org/10.17605/OSF.IO/BNVYK). Data Processing 3D marker trajectories were labeled using QTM software (Qualisys, Göteborg, Sweden), then transferred into Visual3D (C‐Motion Inc., Germantown, USA) to define joint kinematics and kinetics using a standard inverse dynamics approach. The marker trajectories were filtered according to the original study, with a 12 Hz low‐pass fourth order Butterworth filter. Take‐off phase was defined as the period from the maximum knee flexion to the foot taking off the ground. Landing phase was defined as the period from the initial contact on the ground to the maximum knee flexion. The instant of take‐off and landing were determined based on a vertical GRF threshold of 10N. The variables computation followed Lam et al. (2021) description. We therefore calculated jump height (determined as the difference between the maximum height of the midpoint between the PSIS markers during flight and its starting standing position), take‐off velocity and peak GRF, representing the performance variables. We computed also peak joints kinematics and kinetics (i.e., moments and powers) during take‐off and landing phases. Positive value for joint kinematics and kinetics reflected hip flexion, knee extension, ankle dorsiflexion, and MTPJ dorsiflexion. Joint internal moments and powers were normalized by body weight and body height. Moreover, a careful examination of the variables computed in the original study suggests that some variables may not have been reported in accordance with the methods described in the article. Therefore, we modified the metatarsophalangeal joint velocity sign at takeoff (as plantarflexion should be negative). Moreover, the ankle “plantarflexion velocity” and “dorsiflexion moment” at landing were modified as ankle “dorsiflexion velocity” and “plantarflexion moment” at landing, and the signs were revised to be positive in dorsiflexion. Finally, we changed “knee flexion moment” at landing by “knee extension moment” and adapted the sign accordingly. Statistical Analysis For the consecutive jumps condition, only jumps between the second and the fifth were used to compute the different parameters. Then, those were averaged over the six sets. Also, mean values and coefficients of variation (CV) were computed for the single CMJ over the six repetitions and were compared to the consecutive CMJs condition, as per the original study. Normality of data distribution was assessed with a Shapiro‐Wilk test. To identify significant differences between single and consecutive jumps tasks, paired t‐tests or Wilcoxon tests were performed if data was normal or not, respectively. As recommended, we also calculated d z (or r for Wilcoxon signed‐rank test) and the 95% confidence intervals using the MOTE package in R as it is the appropriate effect size estimate for paired designs (Lakens 2013). The effect sizes of the original study were signed to enable the comparison between the two studies. The original study reported applying a Holm–Bonferroni correction to control the false discovery rate of 5%. However, Holm‐Bonferroni procedure controls the family‐wise Type 1 error rate, not the false discovery rate. Inspection of the reported results suggests that some variables should have been significant under a Holm–Bonferroni correction, indicating that a fixed threshold (between 0.002 and 0.003), likely based on a Bonferroni correction, may have been used. Accordingly, to closely replicate the original analysis, the significance level was set at p ≤ 0.002. Original findings were considered as replicated, if results in the replication study had identical statistical significance, in the same direction as the original effect with a similar or larger effect size value (Brandt et al. 2014). After careful replication of the original effect sizes, most of the effect size estimate appears to be expressed as Cohen's d av in the original study, but we were not able to replicate all effect sizes. Statistical comparison was therefore performed with the reported signed effects sizes. All data was analyzed using R (version 4.2.1). Therefore, to evaluate the compatibility of the replication and original effect size estimates, a z‐test was used to determine if the replication and original effect size estimates were statistically different using the TOSTER R package (version 0.5.0), with α ≤ 0.002 (Caldwell 2022). In the case of computation of non‐parametric effects size in one study and parametric effects size in the other study, no statistical comparison was performed. Besides, we compared the original and replication means and CV with an independent t‐test to compare the values of both studies, without focusing on the difference between the conditions. The raw data and code for the replication analyses are available at https://doi.org/10.17605/OSF.IO/BNVYK. Results Comparison of the Jumping Conditions For the performance variables, single jumps induced greater height, take‐off velocity and peak GRF at landing compared to consecutive jumps (p < 0.001; Table 2). MTPJ angles and plantarflexion velocity were significantly lower in consecutive condition during take‐off (p < 0.001). During the take‐off phase, knee extension velocity and power were significantly lower for consecutive jumps than a single jump, but knee extension moment was larger (p < 0.001). The hip presented smaller joint flexion angles, extension velocity, and power in the consecutive jumps, compared to single jumps during take‐off phase (p < 0.002; Table 2). During the landing phase, ankle, knee and hip joints maximal flexion were larger in consecutive condition (p < 0.001). However, ankle angular velocities and power, as well as knee power were greater in single jump condition than consecutive jumps during this landing phase (p < 0.001; Table 2). Jump phases durations were not significantly different between the two jumping conditions. Indeed, take‐off durations were 0.26 ± 0.04s versus 0.27 ± 0.05s for single and consecutive CMJ, respectively, and landing durations were 0.29 ± 0.13s for single versus 0.25 ± 0.4s for consecutive CMJ. TABLE 2Jump performance, ground reaction force, MTPJ, ankle, knee and hip variables during single and consecutive jumps expressed as mean (standard deviation) and CV for the replication study.Single jumpConsecutive jumpsComparison of meansComparison of CVPerformancePhaseMean (SD)CV (%)Mean (SD)CV (%) t/z p ES t/z p ESJump height (m)0.47 (0.06)2.560.42 (0.05)3.1911.23 < 0.001 1.78−1.88r 0.060−0.30r Take‐off velocity (m/s)2.65 (0.22)2.122.57 (0.22)2.644.99 < 0.001 0.79−3.090.004−0.49Peak GRF (N/BW)Take‐off1.22 (0.13)3.871.32 (0.17)4.57−3.88r < 0.001 −0.59r −1.43r 0.154−0.23r Landing2.05 (0.52)20.091.74 (0.25)9.85−4.43r < 0.001 0.65r 9.22 < 0.001 1.46MTPJPeak flexion angle (°)Take‐off9.13 (3.11)16.847.40 (2.32)12.665.48 < 0.001 0.873.49 0.001 0.55Landing9.30 (2.77)25.0010.23 (2.96)12.80−2.390.022−0.38−6.43r < 0.001 0.85r Peak extension velocity (°/s)Take‐off−342.25 (94.64)12.64−293.32 (81.29)11.37−4.73 < 0.001 −0.75−1.070.2920.17Peak flexion velocity (°/s)Landing316.82 (123.83)29.00341.36 (113.86)15.10−1.570.124−0.25−5.30r < 0.001 0.75r AnklePeak dorsiflexion (°)Take‐off29.16 (4.29)4.2729.85 (4.91)3.53−1.960.058−0.312.140.0390.34Landing28.54 (5.38)6.8630.16 (4.66)3.19−4.12 < 0.001 −0.65−5.77r < 0.001 0.80r Peak plantarflexion velocity (°/s)Take‐off−898.34 (109.50)4.03−890.04 (104.29)3.45−1.290.204−0.20−1.670.1020.26Peak dorsiflexion velocity (°/s)Landing989.79 (124.70)6.74902.01 (111.00)4.547.40 < 0.001 1.17−3.08r 0.002 0.47r Peak plantarflexion moment (Nm/BWBH)Take‐off−0.09 (0.01)4.95−0.09 (0.01)6.19−1.91r 0.0570.30r 2.960.005−0.47Landing−0.11 (0.02)17.76−0.10 (0.02)8.34−0.93r 0.354−0.15r −6.47r < 0.001 0.85r Peak joint power (W/BWBH)Take‐off0.57 (0.13)7.520.55 (0.11)5.572.620.0120.413.68 < 0.001 0.58Landing−1.07 (0.30)19.84−0.70 (0.16)12.42−6.82r < 0.001 −0.87r −4.50r < 0.001 0.66r KneePeak flexion angle (°)Take‐off−96.00 (9.16)3.25−96.57 (10.90)2.810.410.6860.06−1.450.1550.23Landing−89.78 (15.52)6.63−96.99 (10.79)2.843.63 < 0.001 0.57−5.22r < 0.001 0.74r Peak extension velocity (°/s)Take‐off920.11 (84.43)3.12834.02 (92.40)3.6512.90 < 0.001 2.04−0.97r 0.334−0.16r Peak flexion velocity (°/s)Landing−617.71 (56.69)6.74−597.91 (55.29)4.01−2.970.005−0.47−5.85r < 0.001 0.80r Peak extension moment (Nm/BWBH)Take‐off0.11 (0.02)6.080.12 (0.02)5.36−4.79 < 0.001 −0.761.400.1680.22Landing0.13 (0.02)14.260.13 (0.01)6.731.570.1240.257.62 < 0.001 1.21Peak joint power (W/BWBH)Take‐off0.76 (0.16)6.070.69 (0.14)4.307.98 < 0.001 1.26−3.95r < 0.001 0.59r Landing−1.16 (0.29)21.87−0.95 (0.19)11.56−6.09r < 0.001 −0.82r −8.18 < 0.001 1.29HipPeak flexion angle (°)Take‐off94.43 (8.14)3.0588.80 (11.57)3.56−3.17r 0.002 0.49r −1.540.132−0.24Landing80.70 (18.30)9.9589.42 (11.63)3.78−3.61 < 0.001 −0.57−4.62r < 0.001 0.69r Peak extension velocity (°/s)Take‐off−584.42 (76.68)4.60−503.89 (66.08)4.78−13.10 < 0.001 −2.13−0.430.670−0.07Peak flexion velocity (°/s)Landing446.95 (90.15)12.64461.82 (70.31)6.81−1.390.171−0.227.12 < 0.001 1.15Peak extension moment (Nm/BWBH)Take‐off−0.12 (0.02)6.84−0.13 (0.03)6.482.410.0210.390.770.4450.13Landing−0.13 (0.04)26.53−0.14 (0.03)9.93−2.84r 0.0040.45r −6.85r < 0.001 0.87r Peak joint power (W/BWBH)Take‐off0.38 (0.09)9.670.35 (0.09)8.313.59 < 0.001 0.57−1.25r 0.2110.20r Landing−0.79 (0.39)34.37−0.77 (0.32)18.29−0.360.723−0.06−5.58 < 0.001 0.79 Note: Bold p‐values refer to significant differences between jump conditions after the Holm‐Bonferroni correction. t/z refers to the t or z value of the statistical test performed. rrefers to the use of a r of Wilcoxon instead of a Cohen's dz when needed. Jump performance, ground reaction force, MTPJ, ankle, knee and hip variables during single and consecutive jumps expressed as mean (standard deviation) and CV for the replication study. Note: Bold p‐values refer to significant differences between jump conditions after the Holm‐Bonferroni correction. t/z refers to the t or z value of the statistical test performed. rrefers to the use of a r of Wilcoxon instead of a Cohen's dz when needed. Comparison of the Means and CV of the Original and the Replication Studies Compared to the original results, the athletes of the current study jumped significantly lower in both conditions (0.58 ± 0.05 m vs. 0.47 ± 0.06 m in single CMJ (p < 0.001) and 0.49 ± 0.05 m versus 0.42 ± 0.05 in consecutive CMJ (p < 0.001)). The CV were mostly smaller in the current study (p ≤ 0.002, for 37 out of 48 variables), compared to the original study. Detailed results comparing the mean and CV values of the two studies were reported in the supplementary materials section. Replication Outcome Effect sizes of the performance variables were in the same direction in both studies, but their statistical differences could not be tested because the computation of the ES was not consistent (non‐parametric vs. parametric) (Figure 1). Regarding the MTPJ, every kinematic and kinetic variable's effect sizes were significantly different between the two studies, except the ES for the peak dorsiflexion angle during landing phase. For the ankle joint, during landing and take‐off, dorsiflexion velocity and joint power effect sizes were in opposite directions in the two studies. The knee variables effect sizes were frequently in opposite direction and/or significantly different (Figure 1). Overall, only 4 variables can be considered as successful replication, 3 as practical failure to replicate, 8 as inconclusive replication and 9 as informative failure to replicate (Table 3). Consequently, the replication and original study findings are mostly incompatible (Brandt et al. 2014). Comparison of performance, MTPJ, ankle and knee joints variables effect sizes between the original study and the replication study. The confidence interval for the MTPJ plantarflexion velocity during takeoff for the original study was not indicated, as the effect size computed based on the original results was different from the one reported. a: significant difference; b: unable to perform statistical comparison. TABLE 3Jump performance, ground reaction force, MTPJ, ankle, knee variables comparison for replication analysis.Replication testOriginal testReplication test outcomeOriginal test outcomeIf both significant, are t‐tests in the same direction?Effect sizes (Z‐test) statistical comparisonConclusionsPerformanceJump height t‐testWilcoxonSignificantSignificantSameCannot computePractical failure to replicateTake‐off velocity t‐testWilcoxonSignificantSignificantSameCannot computePractical failure to replicatePeak GRF take‐offWilcoxonWilcoxonSignificantNon‐significantCompatibleInconclusivePeak GRF landingWilcoxon t‐testSignificantSignificantSameCannot computePractical failure to replicateMTPJPeak flexion angle take‐off t‐test t‐testSignificantNon‐significantSignificant differenceInformative failure to replicatePeak flexion angle landing t‐test t‐testNon‐significantSignificantCompatibleInconclusivePeak extension velocity take‐off t‐test t‐testSignificantSignificantDifferentSignificant differenceInformative failure to replicatePeak extension velocity landing t‐test t‐testNon‐significantSignificantSignificant differenceInformative failure to replicateAnklePeak dorsiflexion angle take‐off t‐test t‐testNon‐significantNon‐significantCompatibleSuccessPeak dorsiflexion angle landing t‐test t‐testSignificantNon‐significantCompatibleInconclusivePeak plantarflexion velocity take‐off t‐testWilcoxonNon‐significantSignificantCannot computeInconclusivePeak dorsiflexion velocity landing t‐test t‐testSignificantSignificantDifferentSignificant differenceInformative failure to replicatePeak plantarflexion moment take‐offWilcoxonWilcoxonNon‐significantNon‐significantCompatibleSuccessPeak plantarflexion moment landingWilcoxonWilcoxonNon‐significantNon‐significantCompatibleSuccessPeak joint power take‐off t‐testWilcoxonNon‐significantSignificantCannot computeInconclusivePeak joint power landingWilcoxonWilcoxonSignificantNon‐significantSignificant differenceInformative failure to replicateKneePeak flexion angle takeoff t‐test t‐testNon‐significantNon‐significantCompatibleSuccessPeak flexion angle landing t‐test t‐testSignificantNon‐significantCompatibleInconclusivePeak extension velocity take‐off t‐testWilcoxonSignificantNon‐significantCannot computeInconclusivePeak flexion velocity landing t‐test t‐testNon‐significantSignificantSignificant differenceInformative failure to replicatePeak extension moment take‐off t‐test t‐testSignificantSignificantDifferentSignificant differenceInformative failure to replicatePeak extension moment landing t‐test t‐testNon‐significantSignificantSignificant differenceInformative failure to replicatePeak joint power take‐off t‐test t‐testSignificantSignificantDifferentSignificant differenceInformative failure to replicatePeak joint power landingWilcoxon t‐testSignificantSignificantDifferentCannot computeInconclusive Note: The replication is considered a success when the outcomes are similar, with compatible effect sizes, a practical failure to replicate when the outcomes are similar with no Z‐test or significantly different effect sizes, inconclusive when the outcomes are different with no Z‐test or compatible effect sizes, and an informative failure to replicate when the significance of the outcome are different with a significant difference in effect sizes. Statistical significance for both the original and replication test outcomes was set at p ≤ 0.002. Jump performance, ground reaction force, MTPJ, ankle, knee variables comparison for replication analysis. Note: The replication is considered a success when the outcomes are similar, with compatible effect sizes, a practical failure to replicate when the outcomes are similar with no Z‐test or significantly different effect sizes, inconclusive when the outcomes are different with no Z‐test or compatible effect sizes, and an informative failure to replicate when the significance of the outcome are different with a significant difference in effect sizes. Statistical significance for both the original and replication test outcomes was set at p ≤ 0.002. Discussion The present study's aims were to evaluate the replicability of the results of Lam et al. (2021) and to determine the biomechanical differences between single and consecutive CMJ. The main result was that we overall couldn't replicate the original findings, as most of the joint kinematics and kinetics results were not consistent between the two studies. Unlike the original study, MTPJ, ankle joint, and knee joint velocities and powers were greater in single CMJ condition compared to consecutive CMJ condition. Additionally, the hip joint displayed larger velocities and power at landing in the single CMJ condition compared to consecutive CMJ condition. During the take‐off phase, determinant for the performance, the present results seem in line with the achieved jump height, as lower angular velocities, lower joint moments (except for the knee), and lower power were observed in the consecutive jumps. The original, and opposite, results (i.e., larger angular velocities, moments and power in consecutive condition) were discussed regarding the lack of information at the hip joint level. As lower angular hip velocities and power were recorded during consecutive jumps compared to the single one, our extension results seem in adequation with the other joints behaviour. Several hypotheses could explain the differences in the two studies. First, the population were different in terms of size, body mass and especially jump height (0.47 ± 0.06 m vs. 0.42 ± 0.05 m). Therefore, the level of expertise of the two populations might not be identical, despite close inclusion criteria and same effect on performance (i.e., a decrease of jump height in consecutive settings). Surprisingly, the larger joint power production in consecutive jumps compared to single jumps in the original study didn't induce higher jumps. Perhaps the push‐off time could have been shorter in the original study, yielding greater GRF and joint power peak values to apply forces within a shorter ground contact time, inducing a lower impulse. While the instructions were identical between both protocols, their application could have differed as the variability of contact time could be large during consecutive countermovement jumps (Bobbert et al. 1986). Moreover, as stated in other studies from Lam and coworkers, the MTPJ passive stiffness, unknown in the replication study, might be correlated to the jumping height in consecutive CMJ (Jia et al. 2022) and could be related to sport's practice (Liu et al. 2020). Also, shoe characteristics could impact jumping performance (Lam et al. 2022) and would accordingly need further description. Overall, we recommend future studies to mention following details in their study, to limit bias in the replication setting: (i) the expertise level of their athletes (at least the number of years of practice, the average number of trainings per week and the level of practice, based on McKay et al. (2022) classification) and (ii) shoe technical characteristics. During landing, basketball athletes displayed larger ankle dorsiflexion and knee flexion angle when performing consecutive jumps, but had lower joint powers, ankle plantarflexion velocity and peak ground reaction forces, compared to single jumps. Therefore, the single CMJ landing induced a stiffer landing, which is consistent with the original study and the literature (Cormack et al. 2008; Lam et al. 2017). However, compared to the original study, our results were inconsistent (opposite effects or same effect with significantly different effect sizes) regarding MTPJ and ankle velocity, ankle power and knee velocity, moment and power. These majors' differences between the two studies can hardly be explained without looking at landing patterns and will be dependent, as mentioned in the precedent discussion section, of how the athletes carried out the task. Besides, discussing the relative injury‐risk based on the results of the original and the current studies would probably lead to overstatements. Indeed, the double‐leg landing in CMJ has never been, to the author's knowledge, a good predictor of any injury, either in single or consecutive conditions. For example, acute injuries like non‐contact ACL ruptures occur mostly in single leg stance (Della Villa et al. 2020), and landing with two legs induced different lower limb mechanics than with one leg landing (Taylor et al. 2016). Moreover, the single‐leg landing has been shown to be associated with future ACL injuries (Dingenen et al. 2015; Numata et al. 2018; Collings et al. 2022) whereas most studies found no evidence with double leg landing tasks (Krosshaug et al. 2016; Nilstad et al. 2021, Leppänen et al., 2017). Furthermore, the landing variables that predicted the future ACL injury in Collings et al.’ study (2022) were obtained during the one‐legged consecutive hop task, and not during the single CMJ. Regarding overuse knee injuries, van der Does et al. (2016) failed to identify kinematic variables predicting the injuries using a CMJ with 75 team sport players, while Verrelst et al. (2014) found that single leg drop jump trunk and hip kinematics were predictive of exertional medial tibial pain. Therefore, while CMJ and/or consecutive CMJ landing assessments may provide useful information, their relevance for injury screening should be interpreted cautiously due to the limited evidence linking these tasks to injury risk. Regarding the inter‐trial variability, the difference of CV between the two jumping tasks was either non‐significant or lower for consecutive jumps, indicating lower variance within participants. However, Lam et al. (2021) reported heterogenous results between variables, and the ankle and knee joints variables presented lower values in our study compared to theirs. All these differences might be explained by the athletic level discrepancies between the two population. First, the original population larger CV might be explained by a better adaptation of their coordination to jump higher, compared to lower‐level athletes (Wilson et al. 2008). Then, the additional constraints induced by the consecutive demand might lead to less adaptative coordination for our population, because of their probable lower athletic level. Overall, in order to improve confidence in the scientific results in sports sciences, we recommend scientists (both authors and reviewers) to be careful of coherence and detailed description the population and of methodological aspects to able close replications. Moreover, it is recommended to adopt Open Science practices (i.e., pre‐study sample size computation, provide raw data and code, and pre‐register the study) to decrease the risk of p‐hacking, underpowered studies, and to increase confidence in the scientific community (Mesquida et al. 2022). Last, we would like to encourage the scientific community in sports sciences to engage themselves in a process of close replication to improve knowledge in our field and enhance the confidence in our results. This replication study, despite being as close as possible of the original study, had some methodological limits. Indeed, the population was a little different from the original one and the shoes weren't the same model. These differences might have impacted our results as discussed above, but we are confident to state that overall, the results of the original study were not replicable and couldn't be generalized to the basketball male population. Additionally, the model used in the original author was not described, therefore marginal differences may occur in sagittal plane kinematics (Josep & Darras, 2023; McCallister & Russel, 2025). These differences are unlikely to affect the interpretation of the results but may slightly alter the raw values. Moreover, further study should investigate these effects on females to enable the generalization of the results. Especially, the female population in sports sciences is under‐investigated (Cowley et al. 2021) and the scientific community should be careful to include much more females in the research protocols. To conclude, the present study couldn't replicate most of the original findings, as the effect of consecutive jump on most of the joint kinematics and kinetics results was not consistent between the two studies. However, we can confirm that consecutive jumps induced lower jumping height and peak GRF at landing, despite having opposite results on joint biomechanics compared to the original study. Within our dataset, hip joint angular velocity and power during take‐off were also lower in consecutive jumps than in the single jump, in agreement with the other joints. Funding This work was supported by the FEDER FSE‐IEJ Lorraine et Massif des Vosges 2014‐2020 under Grant [LO0027294]. Ethics Statement The study was approved by the ethics committee Nord Ouest III (approval reference 2023‐A00362‐43) and was conformed to requirements stipulated in the Declaration of Helsinki. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Supporting Information S1 Table S5: Methodological comparison between the original and replication study. Data Availability Statement The data that support the findings of this study are openly available in OSF at https://doi.org/10.17605/OSF.IO/BNVYK. References Associated Data Supplementary Materials Supporting Information S1 Table S5: Methodological comparison between the original and replication study. Data Availability Statement The data that support the findings of this study are openly available in OSF at https://doi.org/10.17605/OSF.IO/BNVYK.