Predicting Straight Punch Force of Impact

Paul D. House, PhD

University of Central Oklahoma

Jerel L. Cowan, PhD

University of Central Oklahoma


The focus of this study was on the factors that contribute to the impact force of a straight punch. This study consisted of 22 health college aged subjects (23.5 +/- 3.2 years; 17 males, 5 females). There were two testing sessions. During the first session, subjects performed a one arm cable push with their dominant arm. For the second testing session each subject was instructed to punch the bag "as hard as they could". The subjects hit an F-Scan force sensory shoe sole that was used to measure force of impact. The final test was to determine hand speed. A tethered strap from the Humac 360 was looped around each subject's gloved thumb. Again, they were instructed to hit the punching bag "as hard as they could". A multiple regression was the statistical tool used to analyze the data. The independent variables analyzed were: arm length (meters), body weight (kgs), hand speed (meters/second), and 1RM (kgs). The dependent variable was the force of impact measured in newtons. The results of the study showed that model 2, which included 1RM and hand velocity as significant independent variables, was a good predictor of force of impact with r2 = .634. The significant predictors were hand speed and 1RM strength with p = 0.009 and 0.014 respectively. Thus, it would appear that the greatest factors that determine the straight punch force of impact in untrained subjects are hand speed and 1RM strength in a movement pattern that emulates the straight punch.


Historically many forms of martial arts have been developed from various origins throughout the world. Some involve grappling (i.e. wrestling, Jiu Jitsu,Judo, etc.), while others are more striking oriented: boxing, kick boxing, Muey Tai, Karate, Taekwondo, etc. In more contemporary times, there has been a growing interest in combining multiple forms of martial arts in what is referred to as Mixed Martial Arts. The focus of this study was on the striking martial arts, specifically factors that contribute to the impact force of the straight punch.

It has been shown that higher impact forces improve an athlete's chance of success in striking martial arts (Schwartz, et al., 1986; Voigt, 1986). Additionally, investigating the impact force of a strike, the individual's punching power should be considered a relevant factor to the outcome (Neto & Magini, 2008). Since power is the product of force and velocity, then the relative contributions of both strength and speed to impact force are of interest.

When examining the Kung Fu straight palm strike, Neto and Magini (2008) found that experienced Kung Fu practitioners exhibited significantly greater muscle force (p = 0.02), power (p = 0.02) and greater impact force (p = 0.01) and power (p = 0.01) as compared to inexperienced subjects. Of further interest, Neto and Magini (2008) found significant correlation between muscle power and impact power (r2 = .98; p < 0.001) for the experienced group only. This strong correlation was possibly due to a greater striking accuracy from the experienced group. In another study by Neto, Bolander, Pacheco, and Bir (2009), advanced Kung Fu practitioners demonstrated significantly greater (p = 0.017) impact force during maximal punches and palm strikes compared to their intermediate counterparts. Additionally, with both groups combined, the palm strike produced significantly greater (p = 0.028) force on impact compared to the punches.

Impact forces and velocities of movements have been reported in other studies with varying results. For example, Wilk, McNair, and Feld (1983) conducted a study using karate martial artists performing a straight punch into a concrete block. The subjects produced 2400-2800 N of force at velocities ranging from 8.2-10.7 m/s with a mean of 9.5 m/second. Meanwhile, Whiting, Gregor, and Finerman (1988) reported hand velocities of four experienced boxers of 6.6 to 12.5 m/s at the hands just before impact. Gulledge and Dapena (2008) measured the velocity and force of a 3 inch power punch and a reverse punch using 12 expert Karate practitioners. The velocity and force of the 3 inch and reverse punches were 4.09 m/sec with 790 N and 6.43 m/sec with 1450 N respectively. Smith, Dyson, Hale, and Janaway (2000) tested intermediate and experienced boxers' maximal punch force output. They reported impressive values of 3722 N (SD = 133 N) and 4800 N (SD = 227 N) for the intermediate and experienced, respectively. In another study by Smith (2006), Senior England international amateur boxers produced a mean force of 2643 N to the head and 2646 N to the body when using a rear hand straight punch. These were higher values than their lead hand counterparts.

Additionally, Voigt (1989) constructed a dynamometer and tested 10 experienced Karate Martial Artists on peak force and velocity with means of 3334 N (2345-4566 N) and 9.5 m/sec (8.2-10.7 m/s) respectively. Voigt and Klausen (1990) reported an average hand velocity of a straight punch for karate practitioners at 9.35 m/s. Neto, Magini, and Saba (2007) reported experienced Kung Fu subjects' average palm strike velocity was 6.67 m/sec (SD = 1.42) which was significantly faster (p = 0.042) than their inexperienced counterparts 5.04 m/sec (SD = 0.57). Furthermore, Neto et al. (2009) demonstrated that the highest hand velocity during striking occurred just before impact.

There has been other research on velocity of movement during various striking techniques. Piorkowski, Lees, and Barton (2011) reported lead and reverse hooks were significantly faster than the jab and cross punches when measured on 10 experienced martial artists. However, they also reported longer delivery times for the lead and reverse hooks. Of particular interest were the correlations of hand velocities and sum of lower body forces to punch forces of the hook and straight punches reported by Mack, Stojsih, Sherman, Dau, and Bir (2010). They reported r2 values of 0.380 and 0.391 for hand velocities and the forces of the hook and straight punches respectively, p < 0.05. Interestingly, Filimonov, Koptsev, Husyanov, and Nazarov (1983) reported that ground reaction forces contributed more to the strength of a punch in those classified as "Knock Out Artists" compared to those classified as "Players" or "Speedsters" with percentages of 38.6%, 32.8%, and 32.5% respectively. The aforementioned findings further substantiate the importance of lower body strength as a salient contributor to the punching force of impact.

Irineu, Guilherme, Renaldo, Saulo, and Emerson (2014) reported significant correlations (p < 0.05) between punching hand acceleration at a self-selected distance from the target and mean propulsive power during a squat jump (watts/kg) as well as 1 repetition maximum (1RM) bench press with r2 = 0.646 and 0.558 respectively. This indicates that upper and lower body strength and power might be strong predictors of punching impact force.

When measuring impact force of a punch, two other considerations must be taken into account: mass of the object, and arm length. With greater mass there is a potential for increased momentum and kinetic energy which could be important contributing factors for greater impact force. Therefore, it is of interest to consider body mass as a potential factor of punching force. Additionally, since impulse is the product of force and time, with greater arm length comes, potentially, greater time to build force prior to impact. To both authors' knowledge, this is the first study to investigate the correlation between arm length and impact force of a straight punch. Therefore, the purpose of this study was to determine the contributions body weight and arm length have in establishing impact force, as well as arm strength and hand velocity.


After consent to participate was established, body weight, dominant arm length (measured from the most lateral portion of the acromion to the distal head of the 3rd metacarpal), and age were collected for each subject. Arm length was measured and recorded to the nearest ? inch and converted to meters. After this data were compiled there were two testing sessions that followed. The purpose of the first testing session was to determine the maximal weight each subject could lift for one repetition during a one arm cable push. NOTE: the strength was measured with a 1 RM test. The second testing session was conducted in order to find the peak hand speed and force of impact while performing a straight punch. The second testing session was conducted a minimum of 3 days post session one to ensure there was no residual soreness due to the maximal effort one arm cable push testing.


This study consisted of 22 (17 males, 5 females) subjects, of which, 2 were left-handed, and 20 were right handed. The subjects were all healthy college age (23.55 +/- 3.2 years) individuals in which 20 were inexperienced while 2 had some striking experience. All subjects were deemed healthy and able to participate via the Physical Activity Readiness Questionnaire (Shephard, Reading, Thomas, and Gledhill, 2002). Additionally, all subjects gave written consent via signing the informed consent form approved by the Institutional Review Board for The University of Central Oklahoma.


During the first session, the subjects were weighed while in their street clothes, however, shoes, jackets and accessories i.e. keys, wallets, were removed. Following the weight-in, subjects performed the one arm cable push with their dominant arm. During the cable push, subjects' maximal weight they could lift for one repetition (1RM) was established. In order to mimic an actual straight punch, during the 1RM testing subjects stood with their feet staggered leading with their non-dominant side (right handed dominant, left foot lead). Another reason for the staggered/boxing stance during the 1RM testing was a study by Kultury, Verkhoshansk, Filimonov, Husyainov, and Garakyan (1991) showed improvements in shot put distance of 1.35 meters when performed from a "boxers stance" resulting in improved correlations with striking skills. In order to get a true 1RM, there was a padded horizontal cross-bar attached to the weight machine (Promaxima Manufacturing, Houston, TX) that anchored the subjects close to the hip region. Additionally, the height of the cable could be adjusted within 3 inch increments to accommodate each subject's height. The pulley and cable height was set to the approximate shoulder height of each respective subject. Subjects warmed-up with 3 sets of 10 repetitions with the weight set at 15 lbs. Based on their own discretion, subjects increased the weight to begin determining their 1RM. To be considered a successful lift, the subjects had to demonstrate full horizontal adduction and extension of the shoulder and elbow respectively. Subjects were given a minimum of two misses on a weight before the weight was reduced by at least 5 lbs or the prior successful attempt was deemed their 1 RM. Although, the weight plates were in 15 lbs increments, either one or two 5 lbs dumbbells were anchored with Velcro to the top of the weight stack. The 5 lbs increments in weight allowed for a more accurate measure of each subject's maximal one arm cable push. Following the 1RM test, subjects straight punched the punching bag (Original Wavemaster, Century Martial Art Supply LLC, Oklahoma City, OK) 10 times for familiarization purposes. Subjects punched the bag using a Neoprene bag glove (Century Martial Art Supply LLC, Oklahoma City, OK).

For the second testing session, subjects were given 10 warm-up punches from their dominant side staggered stance. The subjects were then positioned to where their outstretched clinched fist was just in reach of the punching bag while in their staggered punching stance. Wearing a Neoprene bag glove (Century Martial Art Supply LLC, Oklahoma City, OK), each subject was instructed to punch the bag (Original Wavemaster, Century Martial Art Supply LLC, Oklahoma City, OK) "as hard as they could". The punches had to hit an F-Scan force sensory shoe sole that was used to measure force of impact in peak newtons (Tekscan, Inc. Boston, MA). The F-Scan was equipped with a 750 hertz sampling rate via the Data Logger system (Tekscan, Inc. Boston, MA). The shoe sole was anchored to the punching bag with adhesive backed Velcro and a Velcro strap (see figure 1). The punching bag was equipped with vertical adjustments capabilities to accommodate each subject's height. The researcher's attempts were to set the height of the force sensor to the approximate shoulder height of each subject. This was to try and replicate the same height as the 1RM testing from session one. Subjects were given a 2 second interval with which they needed to perform the strike. All subjects where given 3 attempts and the highest value was used for analysis.

The final test was to determine hand speed. Each subject was given the same stance, distance, punching bag and glove, and striking instructions as when performing the prior test. A tethered strap from a Humac 360 (Computer Sports Medicine Inc., Stoughton, MA) was looped around each subject's gloved thumb. This was the equipment used to determine hand speed in inches/second. The inches/second was then converted to meters/second. Subjects were given 3 attempts in which their highest value was used for analysis.

Statistical Analysis

A multiple regression with a stepwise progression was the statistical tool used to analyze the data. The independent variables analyzed were: arm length (inches), body weight (lbs), hand speed (meters/second), and 1RM (lbs). The dependent variable was the force of impact measured in peak newtons. To diminish the chance of committing a type II error, the significance level of each predictor variable was set at p = 0.05. Additionally, to assess test-retest reliability for the 3 maximal force and 3 maximal velocity punch trials, intraclass correlation coefficients (ICC) with associated 95% confidence intervals (CI) were used to determine the reliability of scores within the respective trials.


Intraclass correlation coefficients for the force of impact and velocity of hand movement during the punching trials were: 0.921 (95% CI = 0.819-0.969) and 0.9982 (95% CI = 0.960-.993) respectively. These ICCs indicate very strong reliability measures for punch force and velocity. Means and standard deviations for the predictor and criterion variables are found in table 1. The results of the study showed that there were two significant models produced. Model 1 included one predictor variable (hand velocity) which resulted in an f = 19.42; p = 0.000272 and r2 = .493. However, model 2 included two variables (hand velocity and 1RM strength) which yielded the following: f = 16.460; p = 0.00007 as well as r2 = 0.634. This indicates that model 1 and model 2 predict 49.3% and 63.4% of the variance in punching force, respectively. Since model 2 has a higher coefficient of determination, it is a better predictive model.

Table 1. Descriptive Statistics






Std. Deviation

Body Weight (kgs)






Length of dominant arm (meters)






Single arm 1RM cable push (kgs)






Maximum Punch Velocity (meters/Second)






Maximum Punching Force (newtons)






When assessing each individual predictor variable, the strongest predictor was the maximum hand velocity with t = 2.893; p = 0.009 with an unstandardized Beta weight of 51.46. Furthermore, the 1RM cable push was also significance with t = 2.71; p = 0.014 and an unstandardized Beta Weight of 1.116. Following hand velocity and 1RM strength were body weight and arm length respectively. See table 2 for student t tests and significant values as well as Beta weights for each of the predictor variables regressed on the criterion variable, maximal punching force.

Table 2. Model Significance


Unstandardized Coefficients

Standardized Coefficients

t values

p values


Std. Error


Body Weight






Arm Length






1RM _ Cable Push






Maximal Hand Velocity







Of considerable interest was the significance of model 2 in predicting punching force of impact, specifically, the predictive values of hand speed and 1RM on force of impact. These findings are in agreement Mack, et al. (2010), who reported hand velocity had the greatest correlation with punch force for the hook and straight punches followed by the sum of lower body forces. Since the subjects in this study were upright and in a staggered stance, an apparent portion of their force was coming from the lower body and ground reaction forces. These findings also seem to support other research and reviews investigating lower body forces (Filimonov et al., 1983 Lenetsky, Harris, and Brughelli, 2013).

A curious finding was the low force values seen in this study in comparison to most other studies measuring newtons of impact. There are a couple of potential reasons including: inexperienced strikers and the padded/low inertia punching stand used as the striking implement. Although, Neto and Magini (2008) also showed low newtons of forces when experienced Kung Fu practitioners used a palm strike to a basketball. Therefore, the lower the inertia of an object being struck, the less contact time there is between that object and the striking hand potentially diminishing the force of impact. Perhaps, due to this phenomenon, hand speed might be a greater contributive factor to the force of impact than 1RM lifting strength when the target is of low inertia.

Of even greater surprise was that neither body mass nor arm length had any significant predictive value on punching force. Since momentum is the product of mass and velocity, there is a potential for greater momentum with larger masses. However, subjects with greater mass might have had decreased acceleration and final velocity. Additionally, with greater arm length, there could have been a greater potential for impact force since impulse is the product of force and time. Therefore, longer arms could potentially accommodate greater time to build force leading to greater impact force. A possible reason body weight and arm length were not influential in punching force could be due to the fact that inexperienced punchers were used.

The authors suggest that more studies be conducted that include punching forces with experienced striking martial artists. Additionally, the effects of higher inertia striking devices should be employed to gain a deeper understanding of the dynamic relationship of strength and speed on punching forces. Finally, utilizing a variety of different punches as criterion variables would be of interest.


Filimonov, V. I. , Koptsev, K. N. , Husyanov, Z. M. , & Nazarov, S. S. (1983) . Means of
increasing strength of the punch. Journal of Strength and Conditioning Research, 7(6), 65-66.

Gulledge, J. A. , & Dapena, J. K. (2008) . Comparison of the reverse and power punches in
oriental martial arts. Journal of Sports Science 26(2), 189-196.

Irineu, L. , Guilherme, G. A. , Ronaldo, K. , Saulo, G. , & Emerson, F. (2014) . Predicting
punching acceleration from selected strength and power variables in elite karate athletes: A multiple regression analysis.
Journal of Strength and Conditioning Research 28(7), 1826-1832.

Kultury, T. I. P. F. , Verkhoshansk, Y. V. , Filimonov, V. I. , Husyainov, Z. M. , and
Garakyan, A. I. (1991). The dynamics of punching technique and speed-strength in young boxers. SSR, 160-161.

Lenetsky, S. , Harris, N. , & Brughelli, M. (2013) . Assessment and contributions of
punching forces in combat sports athletes: Implications of strength and conditioning.
Strength and Conditioning Journal, 35(2), 1-7.

Mack, J. , Stojsih, S. , Sherman, D. , Dau, N. , and Bir, C. (2010) . Amateur boxer biomechanics and punch force.
28th International Conference on Biomechanics in Sports. Marquette, MI.

Neto, O. P. , Magini, M. , & Saba, M. M. F (2007) . The role of effective mass in the
performance of Kung Fu practitioners compared to non-practitioners. Journal of Applied Biomechanics, 139-148.

Neto, O. P. , and Magini, M (2008) . Comparison of force, power, and striking efficiency
for a Kung Fu strike performed by novice, and experienced practitioners: Preliminary
analysis. Perceptual and Motor Skills 106, 188-196.

Neto, O. P. , Bolander, R. , Pacheco, M. T. T. , & Bir, C. (2009) .Force, reaction time, and
precision of Kung Fu strikes. Perceptual and Motor Skills 109, 295-303.

Piorkowski, B. A. , Lees, A. , & Barton, G. S. (2011) . Single maximal versus combination
punch kinematics. Sports Biomechanics 10(1), 1-11.

Shephard, R. J. , Reading, J. , Thomas, S., & Gledhill, N. (revised 2002). PAR-Q & You
Canadian Home Fitness Test and exercise screening alternatives. Sports Medicine

Smith, M. S. (2006) . Physiological profile of senior and junior england international amateur
boxers. Journal of Sports Science & Medicine, 74-89.

Smith, M. S., Dyson, R.J., Hale, T., and Janaway, L. (2000) . Development of a boxing
dynamometer and its punch force discrimination efficacy. Journal of Sports Sciences, 18,

Schwartz, M. L. , Hudson, A. R. , Fernie, G. R. , Eng, P. , Hayashi, K. , & Coleclough, A. A.
(1986) . Biomechanical study of full-contact karate contrasted with boxing. Journal of Neurosurgery, 64, 248-252

Voigt, M. A. (1989) . Telescoping effect of the human hand and forearm during high energy
impacts. Journal of Biomechanics, 22(10), 1095.

Voigt, M. A. , & Klausen, K. (1990) . Changes in muscle strength and speed of an unloaded
movement after various training programmes. European Journal of Applied Physiology, 60: 370-376.

Whiting, W. C. , Gregor, R. J. , & Finerman, G. A. , (1988) . Kinematic analysis of human
upper extremity movements in boxing. The American Journal of Sports Medicine, 16(2),

Wilk, S. R. , McNair, R. E. , & Feld, M. S. (1983) . The physics of karate. American Journal of
Physics, 51, 783-790.


  • There are currently no refbacks.