Secondary Logo

Journal Logo

Original Research

Lower Limb Maximal Power Predicts Punching Speed in Different Static and Dynamic Attacking Techniques in Karate

Quinzi, Federico; Rosellini, Gioele; Sbriccoli, Paola

Author Information
Journal of Strength and Conditioning Research: May 2022 - Volume 36 - Issue 5 - p 1353-1359
doi: 10.1519/JSC.0000000000003653
  • Free



Muscle power is the ability to express mechanical work per unit of time. In most sport activities, successful performance relies on this ability, among others (9). Literature focusing on combat sports showed that this category of sport is not an exception. Indeed, the importance of muscle power was highlighted in several review articles focusing on taekwondo (2), boxing (8), judo (13), mixed martial arts (17), and karate (6,7). The crucial role of muscle power for successful performance in combat sport was drawn mainly by reviewing studies that compared athletes competing at different level (21,31,34,44), by using correlation analyses (15), or by comparing winners and nonwinners (35) of simulated combat competitions. Noteworthy, most of these studies analyzed unspecific actions such as the squat (34), the countermovement jump (21), or the peak power observed during the back squat or bench press exercises (35). Although these approaches convey valuable information on the adaptations to training, they provide limited information on the contribution of muscle power to the performance of technical actions.

Recently, some studies investigated the relationship between maximal power and selected kinematic variables in attacking techniques of combat sports (4,22,23). It was shown that impact force of the front kick correlated with vertical jump height of the spike jump, which is a vertical jump preceded by 3–4 run-up steps (4). Interestingly, Loturco et al. (23) showed that lower limb power correlated with punch impact force in boxing. Similarly, Loturco et al. (22) showed a linear correlation between maximal lower limb power and punch acceleration in elite karateka. However, in the study of Loturco et al. (22), the relationship between power and punch acceleration was investigated focusing only on one technical action: the gyaku tsuki chudan (punch executed with the back arm pointing at the sternum of the opponent). Moreover, in that study, subjects performed this technical action starting from a static stance, as opposed to their habitual combat competition behavior. Indeed, combat competitions are characterized by a continuous movement of the athletes on the combat area. Such movements are realized with a combination of eccentric and concentric muscle contraction. This combination, also known as stretch-shortening cycle, leads to higher power production compared with a purely concentric contraction, possibly due to the involvement of the series elastic components (9). For these reasons, in dynamic conditions, the observed relationship between maximal lower limb power and punch kinematics may not hold true.

As a consequence, further evidence is needed to establish whether the relationship between lower limb maximal power and punch kinematics in karate holds also for dynamic conditions. In addition, there is a need to confirm that the presence of a relationship between lower limb power and punching performance is not limited to the gyaku tsuki chudan, but it can be generalized also to other punching actions. At practical level, the possibility to apply current knowledge on this relationship to other technical actions and to more ecological conditions, such as dynamic conditions, is of paramount importance because strength and conditioning coaches may predict and monitor the punching performance of their athletes by using a relatively simple, time-saving, and inexpensive assessment of lower limb maximal power.

To this purpose, the first aim of this study is to investigate the relationship between maximal lower limb power during the back squat and punch kinematics in technical actions starting from either static or dynamic conditions. The second am is to establish whether this relationship can be consistently observed across different technical actions, such as the gyaku tsuki chudan and a combination of kizami tsuki and gyaku tsuki, 2 of the most common punching techniques in karate.


Experimental Approach to the Problem

In this study, we used a correlational study design to verify whether there is a relationship between maximal power of upper and lower limbs during a bench press or a back squat and punching speed during 2 typical karate upper limb punching techniques, starting from either static or dynamic conditions. The study was composed by 3 sessions: in the first session, the procedures, aims, and the risks of the study were explained to the subjects. In the same session, they were instructed on the proper execution of the bench press and back squat (familiarization). The remaining 2 sessions, detailed in the following, were performed in random order. In one session, maximal dynamic strength (1 repetition maximum [1RM]) and maximal power of upper and lower limbs were assessed using a triaxial inertial measurement unit (FreePower, Sensorize, Rome, Italy), which was placed according to Picerno et al. (27), during the performance of a bench press and a back squat at the smith machine. Bench press and back squat were selected in agreement with previous literature focusing on karate athletes (22) to evaluate upper and lower limb maximal power, respectively. In the other session, subjects were asked to perform 2 of the most commonly performed upper limb attacking techniques in karate, the gyaku tsuki (GT), and a combination of kizami tsuki and gyaku tsuki (KG). The technical actions selected for this study were performed starting from a static (GTS; KGS) or from a more ecological combat condition in which the attacker was hopping in place, hereinafter referred to as dynamic condition (GTD; KGD). These technical actions were selected due to their large employment in combat competitions (39). For each technical action, the average foot and punch speed were computed and used for further analysis. At single action level, punch speed was used because it has been shown to be sensitive to proficient punching performance (5,42).


Ten healthy male karate practitioners volunteered to participate in this study. The anthropometric characteristics of the subjects are reported in Table 1. The sample size was determined by an a priori sample size calculation based on the results of the study of Loturco et al. (22). The sample size calculation revealed that 10 subjects were sufficient to achieve a statistical power (1 − β) of 0.80 with significance level (α) set at 0.05.

Table 1 - Anthropometric characteristics of the subjects.
Mean SD Min Max
Age (y) 22.3 1.8 20 25
Mass (kg) 71.1 7.3 60 84
Height (m) 1.77 0.04 1.70 1.84
BMI (kg·m2) 22.7 2.0 19.6 25.9

Subjects were eligible to participate in the study if they were ranked at least black belt and had been competing in the Italian national championship in the 2 years preceding the test in kumite. Subjects were excluded from the study if they reported any neuromuscular disease, had a history of upper and lower limb injuries in the 12 months preceding the test, or were involved in regular strength training. Subjects were recruited from the student population of the university. On average, subjects attended 3 training sessions per week each lasting approximately 2 hours. After full explanation of the risks and benefits and of the procedures of the study, subjects signed the institutionally approved informed consent document to participate in the study. This study complies with the declaration of Helsinki on studies with human subjects and was approved by the Institutional Review Board of the University of Rome “Foro Italico” (IRB approval code: CAR 10/2019).


All experimental sessions were video recorded using a full HD video recorder (Sony HandyCam, HDR-CX250E) operating at 50 frames per second. This sampling frequency has been shown to produce valid results for the calculation of movement speed as compared to higher speed video cameras (29) or to gold standard instrumentation (30,37). To record the entire action of the subjects, for all technical actions, the video recorder was positioned on a tripod at one-meter height and at 6-m distance from the subjects, perpendicular to their line of action. An inertial measurement unit (FreePower, Sensorize) was used to record peak acceleration for the computation of maximal power during the bench press and back squat exercises. Bench press and back squat were performed at the smith machine following previous recommendations on the best practices for the assessment of the force-velocity relationship (26) because they allow to standardize the range of movement, and this latter is limited to the vertical axis only (26). Upper and lower extremity testing was separated by 30 minutes at rest. This resting period was granted to avoid possible cross-over effects between upper and lower limbs (19). For each load intensity (40; 60 and 80% of 1RM), subjects were asked to perform 3 repetitions, in agreement with the recommendations of the American Society of Exercise Physiologists (3), with previous literature on the assessment of lower limb power (1,18) and with the vast literature on isokinetic testing (see Refs. 11, 31, among others). The 3 loads were lifted in ascending order. Subjects were instructed to perform each lift at maximal speed. Before starting maximal power assessment, to standardize the range of movement across subjects, the initial and final positions of the limb were determined as required by the FreePower software. Initial and final positions of the 2 exercises were determined after the recommendations of the National Strength and Conditioning Association (36). Trials that did not meet the established range of motion were automatically discarded from the quantification of maximal power. For both exercises, only the concentric phase of the movement was considered. After each lift, subjects were granted 180s rest as recommended in previous literature (26). For each intensity, the mean power output across repetitions was computed as recommended by Picerno (26). This procedure allows to avoid artifacts in the assessment of the maximal power output due to a possible incorrect execution of the lift (26). The 1RM and maximal power of upper and lower limbs were expressed both in absolute and relative (divided by individual mass of the subject) units.

In the other experimental session, subjects were asked to perform 2 of the most common upper limb attacking techniques in karate, the gyaku tsuki (GT), and a combination of kizami tsuki and gyaku tsuki (KG). These technical actions were performed starting from a static (GTS; KGS) or from a dynamic condition (GTD; KGD). The gyaku tsuki is a rear hand punch aiming at the opponent's sternum; its execution entails a forward lunge with the leg contralateral to the arm performing the punch. The kizami tsuki is a front hand punch aiming at the opponent's face or sternum performed with a simultaneous forward lunge with the leg homolateral to the hand performing the punch. In the combination of these punching actions, the gyaku tsuki follows the kizami tsuki. To reproduce the combat conditions, subjects started the technical actions at a self-selected distance from the target (i.e., the distance they would have performed a punching technique in combat competition) and started the technical action in the typical combat position as detailed in previous studies (32,33). They were instructed to perform these actions at maximal speed. The 2 experimental sessions were performed one week apart. Figure 1 depicts the experimental set-up for the assessment of the kinematic variables of the technical actions.

Figure 1.:
Initial (A) and final (B) posture of the subjects in a representative technical action.

Before performing the technical actions, subjects were allowed 10 minutes of self-administered warm-up according to their habitual warm-up routines. For each technical action, 5 attempts, interspersed by one-minute rest, were performed. Only the dominant arm was used. Technical actions were performed on a 0.04-m-thick karate mat, which is approved by the World Karate Federation. The technical actions were performed in random order.

Data Analysis

Kinematic variables of the technical actions were analyzed using the open source software Kinovea (version 0.8.15; Kinovea is a free license software that has been largely used to analyze sporting actions (10,14,30,37). The validity and reliability of this software have been widely investigated showing good to excellent results for its validity (14,29,30,37) and excellent results for its reliability (10,14,30). Before data processing, a calibration grid was placed on the testing area in Kinovea, as recommended by the software package. The calibration grid was carefully aligned to the karate mats where the punching actions were performed. The calibration grid laid over 2 adjacent karate mats covering a 2 × 1-m (length × width) area. Toe-off to heel-contact on the karate mat were identified as the instants of time of foot movement onset and offset, respectively. Punch movement onset and offset were identified as the first frame on the video when the punch started to move toward the target and punch contact with the target (43), respectively. From these discrete events, visually identified on the recorded videos, the duration of foot (ft) and punch (pt) movement was computed. In addition, the linear trajectory of the leading foot (fd) and of the attacking punch (pd) were computed from onset to offset of foot and punch movement, respectively. The total action duration was computed from toe-off to punch movement offset. Mean velocity of punch (pv) and foot (fv) was calculated. For each technical action (gyaku tsuki vs. kizami tsuki and gyaku tsuki) and for each condition (static vs. dynamic), the attempt showing the highest punch velocity was chosen and used for further analysis. In the KG technique, both in the static and dynamic conditions, the kinematic variables were computed on the second punch (gyaku tsuki).

Statistical Analyses

The software Statistica (Statsoft, ver. 10.0) was used to perform the statistical analysis. The normal distribution of the variables of interest was verified by means of the Shapiro-Wilk test. All variables were normally distributed and will be reported in the following paragraphs as mean, SD, and 95% confidence intervals (CIs). To verify the effect of Task (GT vs. KG) and Condition (static vs dynamic) on punch and foot speed and on the total duration of the actions, a 2 × 2 analysis of variance was performed. A one-way analysis of variance was performed to test possible differences between upper and lower limb in maximal dynamic strength (1RM) and peak power both expressed in absolute and relative units. Tukey post hoc analysis was performed where appropriate. Effect size was reported as partial eta square (ηp2). To verify whether there was a relationship between peak power and variables of interest during the technical actions, Pearson's product-moment correlations were performed between the following variables: upper and lower limb power, fv, pv, and total action duration of the 2 technical actions (GT vs. KG) in the 2 conditions (static vs. dynamic). For all statistical analyses, the null hypothesis was rejected with p < 0.05.


In Table 2, peak power and 1RM during the bench press and the back squat exercises are reported in absolute and relative units. In absolute units, lower limbs were stronger (F(1,9) = 27.16; p = 0.0005; ηp2 = 0.75) and more powerful (F(1,9) = 10.62; p = 0.01; ηp2 = 0.54) than upper limbs. When expressed in relative units, the differences between upper and lower limbs were not modified (1RM: F(1,9) = 25.61; p < 0.001; ηp2 = 0.73; Power: F(1,9) = 12.14; p < 0.006; ηp2 = 0.57).

Table 2 - Absolute and relative peak power and maximal dynamic strength (1RM; right) recorded during the bench press (BP) and back squat (BS).*
1RM (kg) 66.3 ± 16.9 (54.2–78.4) 119.1 ± 33.9 (94.9–143.3)
Peak power (W) 301.6 ± 84.6 (241.1–362.1) 436.4 ± 133.9 (340.6–532.2)
Relative 1RM (kg) 0.92 ± 0.21 (0.77–1.07) 1.68 ± 0.48 (1.33–2.03)
Peak relative power (W·kg) 4.20 ± 0.96 (3.50–4.89) 6.08 ± 1.60 (4.93–7.23)
*Data are reported as mean ± SD and 95% confidence intervals.
Significant differences between upper and lower limbs.

The statistical analysis performed on punch speed showed a significant effect of Task (F(1,9) = 10.41: p = 0.01; ηp2 = 0.53). The Tukey post hoc analysis revealed that the combination of kizami tsuki and gyaku tsuki showed higher punch speed (7.00 m·s−1) compared with the execution of the gyaku tsuki alone (6.37 m·s−1p = 0.01). No significant effect of Condition (F(1,9) = 1.45: p = 0.25; ηp2 = 0.13) and no significant Task by Condition interaction (F(1,9) = 0.69: p = 0.42; ηp2 = 0.07) were observed for punch speed (GTS: 6.28 m·s−1; GTD: 6.47 m·s−1; KGS: 6.79 m·s−1; KGD: 7.21 m·s−1). No significant effect of Task (F(1,9) = 1.97: p = 0.19; ηp2 = 0.17), Condition (F(1,9) = 0.80: p = 0.39; ηp2 = 0.08), and Task by Condition interaction (F(1,9) = 0.47: p = 0.50; ηp2 = 0.04) were observed for foot speed (GTS: 3.53 m·s−1; GTD: 3.86 m·s−1; KGS: 4.07 m·s−1; KGD: 4.07 m·s−1). A significant effect of Condition (F(1,9) = 5.31: p = 0.046; ηp2 = 0.37) was observed for the total duration of the action, with the actions starting in dynamic conditions lasting longer (0.35 seconds) than those starting from a static condition (0.32 seconds). No significant effect of Task on total duration was observed (F(1,9) = 4.77: p = 0.056; ηp2 = 0.34) as well as no significant Task by Condition interaction (F(1,9) = 0.32: p = 0.58; ηp2 = 0.03). The results of the statistical analysis of punch and foot speed and on the total action duration are reported in Table 3.

Table 3 - Comparison of kinematics variables across the 2 technical actions executed in the static or dynamic condition.*
Punch speed (m·s−1) 6.28 ± 2.43 (4.54–8.02) 6.69 ± 2.72 (4.84–8.73) 6.47 ± 1.98 (5.05–7.88) 7.21 ± 2.33 (5.54–8.88)
Foot speed (m·s−1) 3.53 ± 1.33 (2.58–4.49) 4.07 ± 1.91 (2.69–5.44) 3.86 ± 1.35 (2.89–4.83) 4.07 ± 1.55 (2.95–5.18)
Total time (s)§ 0.27 ± 0.08 (0.21–0.33) 0.36 ± 0.19 (0.22–0.50) 0.33 ± 0.14 (0.22–0.43) 0.38 ± 0.16 (0.26–0.50)
*GTS = gyaku tsuki static; GTD = gyaku tsuki dynamic; KGS = kizami gyaku static; KGD = kizami gyaku dynamic.
Data are reported as mean ± SD and 95% confidence intervals.
Significant effect of Task.
§Significant effect of Condition.

Significant correlations were observed between lower limb maximal power, expressed in relative units, and punch speed in all considered conditions and tasks (GTS: r = 0.71, 95% CI: 0.26–0.91, p = 0.019; KGS: r = 0.80, 95% CI: 0.44–0.94, p = 0.005; GTD: r = 0.66, 95% CI: 0.05–0.91 p = 0.037; KGD: r = 0.68, 95% CI: 0.08–0.91, p = 0.029; Figure 2, panel A). Conversely, no significant correlations were observed between maximal power of the upper limbs, expressed in relative units, and punch speed (Figure 2B).

Figure 2.:
Correlation analysis between punching speed and relative maximal power recorded during a back squat (BS) and a bench press (BP). Please note that only BS maximal power correlated with punching speed across tasks and conditions. Dashed lines represent 95% confidence intervals.

Significant correlations were observed between lower limb peak power and foot speed across Task and Conditions (all r >0.64 to <0.85 and p < 0.04–0.001). Notably, in both technical actions, larger correlation coefficients were observed when the technical actions were performed in static conditions.


In this study, we investigated the existence of a possible relationship between upper and lower limb maximal power and punch speed during 2 of the most commonly used attacking techniques in karate, which were executed starting either from a static or a dynamic condition. We confirmed previous observations (22) showing significant correlations between lower limb maximal power and punch kinematics in the gyaku tsuki (GTS: r = 0.71; GTD: r = 0.66). Furthermore, we showed that lower limb maximal power can be proficiently used to predict punching speed also in other technical actions such as the kizami gyaku tsuki, which were executed not only starting from a static posture (KGS: r = 0.80) but also in a more ecological dynamic condition (KGD: r = 0.68). Last, since current knowledge on the relationship between lower limb maximal power and punch kinematics (22) or dynamics (23) focused on international level athletes, our results show that previous findings can be successfully applied also to athletes competing at national level.

In agreement with previous literature showing a relationship between lower limb strength and power and punching performance in karate professional athletes (r-values range: 0.77–0.81 (22)) and international-level boxers (r-values range: 0.67–0.85 (23)), in this study, we showed a link between lower limb maximal power and upper limb kinematics (r-values range: 0.66–0.80). In elite boxers, Filimonov et al. (12) showed that lower limbs contributed approximately to 38.5% of the punching force. These results can be explained by considering that punching actions are realized with the contribution of the arms, trunk, and lower limbs (12,40). However, Stanley et al. (38) failed to show significant correlations between lower limb kinetics and peak resultant fist velocity during the jab punch (r = 0.47; n.s.) in national-level boxers. Most of the studies investigating lower limb kinetics or kinematics and punching performance focused on boxing (12,23,38,40,41). However, it has to be noted that punch kinematics may differ between karate and boxing. In fact, while in karate, punching actions entail a quick translation of the attacker towards his opponent, in boxing, most of the time punches are delivered without this forward translation. Although a thorough investigation of the biomechanical and neuromuscular differences in the punching technique between karate and boxing is beyond the scope of this study, some differences may be hypothesized. Possibly, the translation of the body toward the target (i.e., forward translation), as opposed to a static punching action, might imply a different application of the ground reaction forces, mostly relying on its anterior-posterior component as opposed to what observed in boxing where the vertical component is predominant (38). Furthermore, the translation of the body toward the target might alter lower and upper limb coordinative patterns, both in terms of muscle activation and joint kinematics, resulting in a modified functioning of the kinetic chain arising from the lower limb. Possibly, the different biomechanical execution of the punching actions between these combat sports likely accounts for the lack of significant correlations between GRF and punching speed in boxing (38).

In this study, the correlation between lower limb maximal power and punch speed showed larger coefficients in the static condition for both the combination of the kizami Tsuki and gyaku tsuki and for the gyaku tsuki alone (KGS: r = 0.80; GTS: r = 0.71) than in the dynamic condition (GTD: r = 0.66; KGD: r = 0.68). Previous observations focusing on national- and international-level karateka (34,35) showed that successful performance in karate is associated with higher jumping ability in the squat jump but not in the countermovement jump (34,35), supporting the hypothesis that skilled karate athletes rely more on the ability to express high gradients of force when they start from a static condition rather than from a stretch-shortening contraction. An alternative interpretation could be found in the commonalities between the back squat and the execution of the technical actions in the static condition. Indeed, the assessment of maximal lower limb power was performed with the subjects starting from a static position, thus avoiding any possible countermovement before the execution of the back squat, similarly to what was observed in the technical actions in the static condition. In our opinion, this similarity may account for the higher correlation coefficients between lower limb power and punch speed in static conditions, more than the specific ability to rely on different contraction modality.

Punch speed in this study (range: 6.28–7.21 m·s−1) is in line with previous literature focusing on punching velocity in boxing and kickboxing amateur practitioners (20,28,38), muay thai regional champions (40), expert kung fu practitioners (25), elite karateka (5), and expert martial artists (16), (punch speed range: 6.6–8.2 m·s−1) but see the seminal study of Vos and Binkhorst on 3 national-level karateka (42) for peak punching speeds up to 13.5 m·s−1. Noteworthy, although some of these studies used more accurate instrumentation for the continuous recording of the position of body segments, our results on punch speed are comparable with most of these studies. As a final remark, it is worth noting that in this study, the combination of kizami tsuki and gyaku tsuki resulted in higher punch speed compared with the gyaku tsuki alone (KGS: 6.69 m·s−1; GTS: 6.28 m·s−1; KGD: 7.21 m·s−1; GTD: 6.47 m·s−1). Because both arms contribute to the execution of this combination of attacking techniques, it is conceivable that the execution of the first punch (i.e., the kizami tsuki) may be associated with a rotation of the trunk in a direction opposite to the limb performing the kizami tsuki, which occurs immediately before the execution of the gyaku tsuki with the contralateral limb. The rotation of the trunk may in turn favor the stretch-shortening contraction of trunk muscles contributing to the higher punching speed in the combination of kizami tsuki and gyaku tsuki compared with the gyaku tsuki alone. This interpretation would be in agreement with the study of Tong-Iam et al. (40), in which the role of trunk rotation in performing forceful punches was highlighted in boxing.

The results presented in this study shall be interpreted acknowledging some limitations: first, in this study, only male, high-level, karate subjects were recruited. This aspect might limit the possibility of generalizing the present results to female practitioners or athletes with different training backgrounds (i.e., lower training experience), since this experiment was performed with black belt karate athletes competing at national level. The second limitation concerns the sampling frequency of the video recording system. Indeed, sampling frequencies higher than those used in this study would have enabled a more accurate quantification of the duration of the technical actions. Notwithstanding this issue, it is worth noting that punching speeds computed in this study fall within the range reported in previous observations. Future studies are recommended to verify if, as previously observed in elite boxers (24), significant increases in lower limb maximal power may be transferred to punching performance in elite karate athletes.

Practical Applications

Previous evidence showed that, in single actions performed starting from static condition, punching performance of karate can be predicted from lower limb maximal power. In this study, we aimed at verifying whether this possibility holds true also for more ecological conditions, such as those in which athletes start their action from a dynamic condition and for a combination of 2 techniques (kizami tsuki and gyaku tsuki). We showed that punching speed can be proficiently predicted by lower limb maximal power also in such ecological conditions. Most importantly, this study showed that lower limb maximal power is a good and consistent predictor of punching speed across different punching actions in karate. Karate masters and coaches are strongly encouraged to assess lower limb maximal relative power of their athletes to predict and monitor, using the equations provided in this study (GTS punch speed = −0.34 + 1.08·relative back squat power; GTD punch speed = 1.50 + 0.81·relative back squat power; KGS punch speed = 1.47 + 1.35·relative back squat power; KGD punch speed = 1.15 + 0.99·relative back squat power), the punching performance of their athletes in 2 of the most commonly performed punching actions of karate combat competitions. Because lower limb maximal power assessment is relatively simple, time saving, and inexpensive compared with video analysis or to more accurate motion analysis systems, the present results may represent an efficient tool for coaches to monitor the performance of their athletes and to evaluate the effectiveness of their training programs.


The authors declare conflict of interests. This work was performed in the framework of the “Inclusive Karate: a new perspective to decrease sedentary lifestyle and increase self-confidence in Down Syndrome—IKONS” project cofunded by the Erasmus + Program of the European Union (G.A. 2018-2512). The European Commission support for the production of this article does not constitute an endorsement of the contents, which reflects the views only of the authors, and the Commission cannot be held responsible for any use, which may be made of the information contained therein. The results of this study do not constitute endorsement of the product by the authors or the NSCA.


1. Bosquet L, Porta-Benache J, Blais J. Validity of a commercial linear encoder to estimate bench press 1 RM from the force-velocity relationship. J Sports Sci Med 9: 459–463, 2010.
2. Bridge CA, Ferreira J. Physical and physiological profiles of taekwondo athletes. Sports Med 44: 713–733, 2014.
3. Brown LEEE, Weir JP. ASEP Procedures recommendation I: Accurate assessment of muscular strength and power. J Exerc Physiol Online 4: 1–21, 2001.
4. Buśko K, Nikolaidis PT. Biomechanical characteristics of Taekwondo athletes: Kicks and punches vs. laboratory tests. Biomed Hum Kinet 10: 81–88, 2018.
5. Cesari P, Bertucco M. Coupling between punch efficacy and body stability for elite karate. J Sci Med Sport 11: 353–356, 2008.
6. Chaabène H, Hachana Y, Franchini E, Mkaouer B, Chamari K. Physical and physiological profile of elite karate athletes. Sports Med 42: 829–843, 2012.
7. Chaabene H, Negra Y, Capranica L, Prieske O, Granacher U. A needs analysis of karate kumite with recommendations for performance testing and training. Strength Cond J 41: 35–64, 2019.
8. Chaabene H, Tabben M, Mkaouer B, et al. Amateur boxing : Physical and physiological attributes. Sports Med 45: 337–352, 2015.
9. Cormie P, McGuigan MR, Newton RU. Developing maximal neuromuscular power: Part 2: training considerations for improving maximal power production. Sports Med 41: 125–146, 2011.
10. Dingenen B, Barton C, Janssen T, Benoit A, Malliaras P. Physical Therapy in Sport Test-retest reliability of two-dimensional video analysis during running. Phys Ther Sport 33: 40–47, 2018.
11. Felici F, Bazzucchi I, Sgrò P, et al. Acute severe male hypo-testosteronemia affects central motor command in humans. J Electromyogr Kinesiol 28: 184–192, 2016.
12. Filimonov V, Koptsev KN, Husyanov ZM, Nazarov S. Means of increasing strength of the punch. NSCA J 7: 65–66, 1985.
13. Franchini E, Del Vecchio FB, Matsushigue KA, Artioli GG. Physiological profiles of elite judo athletes. Sports Med 41: 147–166, 2011.
14. Grigg J, Haakonssen E, Rathbone E, et al. The validity and intra-tester reliability of markerless motion capture to analyse kinematics of the BMX Supercross gate start. Sports Biomech 17: 383–401, 2018.
15. Guidetti L, Musulin A, Baldari C. Physiological factors in middleweight boxing performance. J Sports Med Phys Fitness 42: 309–314, 2002.
16. Gulledge KJ, Dapena J. A comparison of the reverse and power punches in oriental martial arts. J Sports Sci 26: 189–196, 2008.
17. James LP, Haff GG, Kelly VG, Beckman EM. Towards a determination of the physiological characteristics distinguishing successful mixed martial arts athletes : A systematic review of combat sport literature. Sports Med 46: 1525–1551, 2016.
18. Jidovtseff B, Harris NK, Crielaard J-M, Cronin JB. Using the load-velocity relationship for 1-RM prediction. J Strength Cond Res 25: 267–270, 2011.
19. Kennedy A, Hug F, Sveistrup H, Guével A. Fatiguing handgrip exercise alters maximal force-generating capacity of plantar-flexors. Eur J Appl Physiol 113: 559–566, 2013.
20. Kimm D, Thiel DV. Hand speed measurements in boxing. Proced Eng 112: 502–506, 2015.
21. Koropanovski N, Berjan B, Bozic PR, et al. Anthropometric and physical performance profiles of elite karate kumite and kata competitors. J Hum Kinet 30: 107–114, 2011.
22. Loturco I, Artioli GG, Kobal R, Gil S, Franchini E. Predicting punch acceleration from selected strength and power variables in elite karate athletes: A multiple regression analysis. J Strength Cond Res 28: 1826–1832, 2014.
23. Loturco I, Nakamura FY, Artioli GG, et al. Strength and power qualities are highly associated with punching impact in amateur boxers. J Strength Cond Res 30: 109–116, 2016.
24. Loturco I, Pereira LA, Kobal R, et al. Transference effect of short-term optimum power load training on the punching impact of elite boxers. J Strength Cond Res 35: 2373–2378, 2019.
25. Neto OP, Magini M, Pacheco MTT. Electromyographic study of a sequence of Yau-Man Kung Fu palm strikes with and without impact. J Sports Sci Med 6: 23–27, 2007.
26. Picerno P. Good practice rules for the assessment of the force-velocity relationship in isoinertial resistance exercises. Asian J Sports Med 8: e15590, 2017.
27. Picerno P, Iannetta D, Comotto S, et al. 1RM prediction: A novel methodology based on the force–velocity and load–velocity relationships. Eur J Appl Physiol 116: 2035–2043, 2016.
28. Piorkowski BA, Lees A, Barton GJ. Single maximal versus combination punch kinematics. Sports Biomech 10: 1–11, 2011.
29. Post A, Koncan D, Kendall M, et al. Analysis of speed accuracy using video analysis software. Sport Eng 21: 235–241, 2018.
30. Puig-Divì A, Escalona-Marfil C, Padullés-Riu J, et al. Validity and reliability of the Kinovea program in obtaining angles and distances using coordinates in 4 perspectives. PLoS One 6: 1–14, 2019.
31. Quinzi F, Bianchetti A, Felici F, Sbriccoli P. Higher torque and muscle fibre conduction velocity of the biceps brachii in karate practitioners during isokinetic contractions. J Electromyogr Kinesiol 40: 81–87, 2018.
32. Quinzi F, Camomilla V, Felici F, Di Mario A, Sbriccoli P. Agonist and antagonist muscle activation in elite athletes: Influence of age. Eur J Appl Physiol 115: 47–56, 2015.
33. Quinzi F, Camomilla V, Di Mario A, Felici F, Sbriccoli P. Repeated kicking actions in karate: Effect on technical execution in elite practitioners. Int J Sports Physiol Perform 11: 363–369, 2016.
34. Ravier G, Grappe F, Rouillon JD. Application of the force-velocity cycle ergometer test and vertical jump test in the functional assessment of karate competitor. J Sports Med Phys Fitness 44: 349–355, 2004.
35. Roschel H, Batista M, Monteiro R, et al. Association between neuromuscular tests and kumite performance on the Brazilian Karate National Team. J Sports Sci Med 8: 20–24, 2009.
36. Sands WA, Wurth JJ, Hewit JK. Exercise technique (Chapter 4). In: The National Strength and Conditioning Association's (NSCA) Basics of Strength and Conditioning Manual. Colorado Springs, CO: NSCA 2012. pp. 48–53.
37. Schurr SA, Marshall AN, Resch JE, Saliba SA. Two dimensional video analysis is comparable to 3d motion in lower extremity movement assessment. Int J Sports Phys Ther 12: 163–172, 2017.
38. Stanley E, Thomson E, Smith G, Lamb KL. An analysis of the three-dimensional kinetics and kinematics of maximal effort punches among amateur boxers. Int J Perform Anal Sport 18: 835–854, 2018.
39. Tabben M, Chaouachi A, Mahfoudhi M, et al. Physical and physiological characteristics of high-level combat sport athletes. J Combat Sport Martial Arts 5: 1–5, 2014.
40. Tong-Iam R, Rachanavy P, Lawsirirat C. Kinematic and kinetic analysis of throwing a straight punch: The role of trunk rotation in delivering a powerful straight punch. J Phys Educ Sport 17: 2538–2543, 2017.
41. Turner A, Baker E, Miller S. Increasing the impact force of the rear hand punch. Strength Cond J 33: 2–9, 2011.
42. Vos JA, Binkhorst RA. Velocity and Force of some karate arm-movements. Nature 211: 89–90, 1966.
43. Whiting WC, Gregor RJ, Finerman GA. Kinematic analysis of human upper extremity movements in boxing. Am J Sports Med 16: 130–136, 1987.
44. Zehr EP, Sale DG, Dowling JJ. Ballistic movement performance in karate athletes. Med Sci Sports Exerc 29: 1366–1373, 1997.

1RM; punching performance; combat sports; kinematics

© 2020 National Strength and Conditioning Association