Secondary Logo

Journal Logo

Original Research

The Three-Dimensional Kinematics of a Barbell During the Snatch of Taiwanese Weightlifters

Chiu, Hung-Ta; Wang, Chih-Hung; Cheng, Kuangyou B

Author Information
Journal of Strength and Conditioning Research: June 2010 - Volume 24 - Issue 6 - p 1520-1526
doi: 10.1519/JSC.0b013e3181db23f4
  • Free



Olympic weightlifting is a sporting event that requires high technique and stability. In the snatch, the lifter pulls the barbell from the platform and catches it overhead in a continuous motion with arms fully extended, and then stands with the barbell in control. In a real competition, a failed attempt usually occurs when the barbell falls in front of or behind the weightlifter during the catch phase because of an incorrect amount of pull force or a poor catching technique.

The general kinematic characteristics of barbell trajectories during the snatch for elite weightlifters have been investigated in previous studies (1,3,5,7). During a snatch, the barbell moves toward the lifter after being lifted off the platform, and then it is pushed away from the lifter's body by the lifter's hip and knee joint extensions and shoulder flexions. Finally, the lifter quickly squats down under the bar readying himself to catch the bar resulting in the barbell moving toward his body again. Different barbell pathways have been found among lifters from different countries: for example, U.S. lifters are more likely to pull the bar forward and to catch the bar forward (9), whereas the most elite Asian lifters pull the bar with a jump backwards resulting in a catch position behind the initial position of the barbell (5,7).

In previous studies, the horizontal travel range (the horizontal displacement from pushing the bar away from the body to the most forward position), the maximum height, the catch position, the vertical travel range (the vertical displacement from the maximum height to the catch position), and the maximum vertical velocity of the barbell have been chosen to be the parameters to characterize a barbell's trajectory. In comparisons between successful and failed lifts, no significant differences have been found in most of the above parameters (9). However, these parameters do not fully represent the overall characteristics of a trajectory of a barbell. From a training viewpoint, these acquired parameters cannot help lifters improve their performance.

Two-dimensional kinematic analysis has often been used to study snatch movements (6-8,11). In these studies, only one camera was set to capture the movement of the barbell on the sagittal plane. During each lift in real competitions, the barbell probably will not move in the same plane because of their different initial positions on the platform. Therefore, using one camera to capture the kinematics of a barbell moving in a different plane will result in measurement errors. Gourgoulis (5) used 2 cameras that were positioned in front of the lifter to film the 3-dimensional movements of a barbell and a lifter in competition. However, the trajectory of the barbell was determined by the inside position of the lifter's hand.

The joint motion of a lifter during the snatch will also influence the trajectory of the barbell. Most previous studies analyzed the kinematics of the lower limb joints to provide information on successful or unsuccessful lifts (1,5). However, there are no “successful” body motions for the lifters described in these studies despite the remarkable differences in the parameters for body motion that have been found between elite lifters (1). From the view point of training, it is difficult to improve on a snatch technique by instructing the lifters to change their joint angles or their joint angular velocities. Monitoring the trajectory of a barbell is perhaps an easier approach to evaluate the performance of the lifter. Therefore, this study attempts to characterize the trajectory of a barbell and to clarify whether there is a standard pattern in the barbell trajectory for each lifter.


Experimental Approach to the Problem

Two high speed cameras are used in this study to film the barbell trajectories for Taiwanese weightlifters under competitive conditions. The characteristics of the barbell's movements for different levels of weightlifters were compared using a statistical approach. This experiment was approved by the Chinese Taipei Weightlifting Association and the experimental procedures did not interfere with the weightlifters' performances during any part of the competition process.


The subjects included 19 male Taiwanese weightlifters (age: 22.3 ± 3.6 years, body mass: 83.9 ± 16.8 kg, height: 169.4 ± 6.7 cm) who participated in the 2006 Asian Games qualifier on the Taiwan National Team. The physical characteristics of the subjects are described in Table 1. It is noteworthy that subjects 1-7 have more than 10 years of training in weightlifting and that subject 1 has won gold medals in international weightlifting competitions. Twenty-four successful lifts were filmed in this study. These lifts are classified into 3 groups by relative barbell-mass (RBM, the ratio of the barbell mass to the lifter's body weight). Figure 1 shows the distribution of the RBM in the successful lifts in each weight category. The 3 groups are assigned to be the better-performance group (RBM > 1.63), the middle group (1.28 < RBM < 1.63), and the worse-performance group (RBM < 1.28). There are 8 lifts in each group. T1 represents the best lift (RBM = 2.09), whereas T24 represents the worst lift (RBM = 0.92). This investigation was approved by the Human Experiment and Ethics Committee of National Cheng Kung University Hospital. All subjects in this study were informed of the experimental risks and signed an informed consent before their participation.

Table 1
Table 1:
Characteristics of weightlifters and the barbell.
Figure 1
Figure 1:
The relative barbell-mass distributions of the 24 successful lifts in this study. There are 8 lifts in each group.


Two high-speed cameras (Mega-speed MS1000, sampling rate = 120 Hz) were set on the left side of the lifter to film the trajectory of the left barbell. The angle between the optical axes of the 2 cameras was about 90° (Figure 2). A calibration cubic volume, 148 cm long, 100 cm wide, and 175 cm high, with 24 control points was used in this study. The 3-dimensional spatial coordinates of the selected points were calculated using the direct linear transformation procedure of the Kwon3D motion analysis software. The raw data were smoothed using a fourth-order Butterworth low-pass filter with a cut frequency of 6 Hz. When the lifter was ready to start to lift the barbell off the platform, an experienced experimenter pushed a button triggering 2 synchronized cameras to film the snatch motion.

Figure 2
Figure 2:
An illustration of the experimental set-up.

This analysis focuses on the snatch movement from the beginning of the barbell lift-off to the instant when the lifter catches the barbell overhead. In previous studies, the snatch was divided into 5 phases: the first pull, the transition, the second pull, the turnover under the barbell, and the catch phase. These phases are primarily determined by changes in the lifter's knee angle. However, because the athletes lift the barbell as close to their bodies as possible to decrease the torque produced by the barbells' weight, 5 new events were chosen for this study. To easily describe the barbell's trajectory, the 5 events are defined as the barbell lifting off the floor (LO), the barbell clearing the knee of the lifter (CK), the lifter extending his hip joints to push the bar away from his body (PB), the barbell reaching its maximum vertical height (MH), and the lifter catching the bar overhead (CB). To compare between lifters, the vertical positions are normalized by the height of the lowest position of the barbell during each catch phase.

Six horizontal variables, including the horizontal position of the barbell at CK (XCK) and PB (XPB), the horizontal travel range (HTR, the horizontal displacement measured starting from the barbell being pushed away from the body to its most forward position), the projectile range (PR, the horizontal displacement from the barbell's most forward position to the catch position), the horizontal velocity of the barbell at its highest position (MHVh), and the maximum horizontal velocity after the lifter pushes the barbell away (MaxVh), and 4 vertical variables, including the maximum height (MH), the normalized maximum height (N-MH, the maximum height normalized by the height of the catch position), the vertical travel range (VTR, the vertical displacement from the maximum height to the catch position), and the maximum vertical velocity (MaxVV) of the barbell, are defined and measured for each lift.

Statistical Analyses

Two experienced experimenters digitized the coordinates that the barbell to make sure of the data are reliable. The test-retest reliabilities are determined using an intraclass correlation coefficient (ICC). A coefficient below 0.40 is considered poor, 0.40-0.59 fair, 0.60-0.74 good, and 0.75-1.00 excellent (2). This 1-way ANOVA is used to evaluate the differences in the 10 barbell kinematical variables during the snatch for each of the 3 groups. The LSD method is used for post hoc comparisons and the statistical significance level is set at p ≤ 0.05. All statistical calculations were performed using the SPSS version 13.0 for Windows (SPSS, Inc., Chicago, IL, USA).


Table 2 shows the ICCs for all the dependent variables. The high coefficients (0.72-0.99) indicate that the test-retest reliabilities are good-to-excellent for all the variables measured in this study.

Table 2
Table 2:
The intraclass correlation coefficients for test-retest reliability of all dependent variables.

The very small displacement of the barbell in the media-lateral horizontal direction shows that the barbell movement is almost exclusively in the sagittal plane for all the lifters. The 3 horizontal and the 3 vertical variables are significantly different in all of the 3 groups (Table 3). XCK is significantly greater in the middle and better groups as compared with the worse group. A smaller PR and a slower MHVh are observed for the better group as compared with the worse group. Although a smaller MH is shown for the middle and better groups, a greater normalized MH is found for the better group as opposed to the other 2 groups. The better group also has a greater vertical travel range than the middle and worse groups.

Table 3
Table 3:
The kinematic variables of a barbell for the 3 groups (mean ±SD, n = 8 per group).

The comparisons in the barbell trajectories of the 4 best lifts (RBM > 1.91) and the 4 worst lifts (RBM < 1.21) are illustrated in Figure 3. The paths of the barbells are similar until the barbell achieves a height of about 50 cm (before the PB event) for T1 (RBM = 2.09), T2 (RBM = 1.99), and T4 (RBM = 1.91). And with the exception of T3 (RBM = 1.97), the paths of the other 3 lifts stay to the right of the vertical reference line. So, compared with the 4 best lifts, the paths of the 4 worst lifts all cross the vertical reference line.

Figure 3
Figure 3:
The barbell trajectories of the 4 best lifts (left figure) and the 4 worst lifts (right figure). The weightlifter stands on the right of the vertical reference line that is drawn through the center of the barbell just before the lift. Compared with the 4 best lifts (except for T3) the paths of all 4 of the worst lifts cross the vertical reference line.

Phase diagrams showing displacement-velocity relationships are used in this study to express the movement patterns of the barbell (Figure 4). The results of the movement patterns show that there is greater variation in the horizontal movement of the barbell than in the vertical movement. The greater horizontal displacement and velocity of the barbell is shown for the 4 worst lifts and than for the 4 best lifts. The movements in the vertical direction show that a greater normalized maximum height and a greater vertical travel range occur in the 4 best lifts. The 4 best lifts seem to show the same pattern for the first 60% in height starting from the lowest position of the barbell during the catch phase.

Figure 4
Figure 4:
The movement patterns of a barbell in the (A) horizontal and (B) vertical directions for the 4 best lifts (left figures) and the 4 worst lifts (right figures). Greater horizontal displacement and velocity, smaller normalized maximum height, and vertical travel ranges of the barbell are shown for the 4 worst lifts and compared with the 4 best lifts.


The purpose of this study is to describe the movement of the barbell during the snatch phase using a 3-dimensional kinematic approach. A very small displacement of the barbell in the media-lateral horizontal direction was found in this study. This shows that 2-dimensional kinematics in the sagittal plane is appropriate for describing a barbell's movement in weightlifting. This is not true, however, if the initial position of the barbell is different for each lift in real competition conditions, for example. Therefore, in this study a 3-dimensional kinematic analysis is necessary to decrease the measurement errors that will occur in the 2-dimensional approach.

In the study of Thé and Ploutz-Snyder (10), a prediction equation for weightlifting performance was derived from cross-validation analyses dependent on the age and body mass of the lifter as follows: ABS (kg) = 259.854 + (−3.147*age [yr]) + (1.286*body mass [kg]), where the ABS is the barbell mass summation in the snatch and clear and jerk. In the present study, the average of the actual achievements in the better group is significantly greater than their respective predicted accomplishments by about 28.2 ± 14.6 kg. It is worth to note that the mass summation of subject 1, the lifter performing at level T1 in this competition, is greater than his predicted value by about 45.6 kg. This champion lifter has won gold medals in international weightlifting competitions in his 62 kg weight class. The RBM indexes of the lifts for the better group, more than 1.63, in this study are similar to the indexes for elite Asian lifters (5,7). This indicates that the better group in this study which is determined by their RBM is representative of elite lifters.

The horizontal variables of the barbell have shown greater variation between lifters than the vertical variables (5). Also, less variation in the horizontal movement of barbell occurs more often for the best lifts than for the poorest lifts (1). In this study, although only XCK, PR, and MHVH are significantly different between groups, even from the worse group to the better group all the mean horizontal variables systematically increased for XCK and XPB or decreased as is the case for HTR, PR, MaxVH, and MHVH. The lack of a statistically significant difference is perhaps because of smaller statistical power values rejecting the null hypotheses. The results indicate that, when just starting to lift the barbell off the platform, a barbell pulled more toward the lifter will result in a better lift. After this, a barbell from a better lift moves with less horizontal displacement and velocity even after being pushed away from a lifter. As shown in Figure 3, the barbell trajectories of the best lifts, except for T3, do not cross the vertical reference line because of the first pull motion being toward the lifter and horizontal travel ranges being less than XPB. This barbell trajectory has been used by elite male Asian lifters for the snatch, and is considered to be a new and widely used technique (1,7).

The vertical variables of a barbell have often been used to evaluate the techniques of lifers. Elite lifters have been characterized by steady increases in the velocity of the barbell and by the formation of a single peak in the vertical velocity curve (1,5,7). In this study, no notable dip was observed in the best or worst lifts, except for T23 (Figure 4). The maximum height of the barbell is higher for the worse group because of higher lifters in the worse group. The lifters' heights are 165, 170, and 175 cm for the better, middle, and worse groups, respectively. Interestingly, a greater mean normalized maximum height and vertical travel range is found in the better group. Minimization of the vertical travel range of the barbell had been considered to be important for an effective technique (7). However, a vertical travel range is necessary to permit the lifter to drop under the bar. In this study, the skillful lifters pull the bar to a higher position to attain a greater vertical travel range. This allows for a gradual slowing down of a barbell's drop velocity because of gravity. With this catching technique, a lifter can catch heavier barbells successfully with the same amount of muscle strength.

In this study, the mean vertical travel range of the barbell for the better group 19.5 ± 1.9 cm is greater than the results of male weightlifters in previous studies which show only a range of about 10-14 cm (1,5,7). Similar ranges of about 19-21 cm (4,6) have been found for female weightlifters. For male weightlifters, the vertical travel range is increased as the barbell mass is increased (1,7). From the above results, the lifters seem to increase their vertical drop displacement of a barbell to catch the barbell under weaker strength or heavier barbell mass conditions. The best lifters in this study are good at using this cushioning technique to catch a heavier barbell successfully.

There has not been any kinematic variables of a barbell that clearly show separate successful lifts from unsuccessful lifts by statistical methods (9). Variability also occurs in the horizontal variables in this study, but there are no significant differences between the variables of successful and unsuccessful lifts. Therefore, after surveying the movement patterns of a barbell it is better to evaluate the snatch technique rather than just to capture the kinematic variables. The displacement-velocity relationship curve forms a circle from the time the barbell clears the knee until it reaches a maximum positive velocity in the horizontal plane. To minimize the area of the indicated circle, it is an effective lifting technique that the barbell should move nearly parallel along the vertical reference line in order to reduce horizontal work (7), for example, in the curve of T1 (Figure 4). When looking at the vertical component, the same pattern has been observed for the 4 best and worst lifts, with the exception of a smaller vertical displacement from the maximum height to the catch position for the worst lifts which we discussed in detail above.

From the results of this study, we conclude that elite Taiwanese weightlifters pull the barbell more toward their bodies from the time the barbell lifts off the platform until it is pushed away from the lifters with complete extension of the hip and knee joints. Then they jump backwards to catch the barbell leading to a greater vertical travel range. The horizontal movement patterns of the barbell show that elite Taiwanese weightlifters perform with less displacement and velocity in the horizontal direction during the pull phase. Although the movement pattern in the vertical direction is similar, there is no standard trajectory for each weightlifter because of the variations in their horizontal movement patterns.

Practical Applications

Asian male weightlifters usually perform well in world weightlifting competitions. In this study, elite Taiwan male lifters perform the same barbell trajectories as other Asian male lifters, except that they have an increased vertical travel range. This catching technique reveals a higher maximum height for the barbell and a backwards jump to catch the barbell in a deeper squat position. The weightlifters seem to have been instructed to increase the vertical travel range of the barbell in order to catch it under weaker strength (e.g. in female lifters) or heavier barbell mass conditions. The results of this study suggest that coaches can film the trajectory of a barbell during a lifter's snatch and replay it immediately to give real time feedback to the lifter and instruct him or her how to find a more correct and successful barbell trajectory.


The authors would like to thank the National Science Council in Taiwan for their financial support of this project (NSC-95-2413-H-006-014). We would also like to thank the Chinese Taipei Weightlifting Association for its approval of this experiment.


1. Baumann, W, Gross, V, Quade, K, Galbierz, P, and Schwirtz, A. The snatch technique of world class weightlifters at the 1985 World Championships. Int J Sport Biomech 4: 68-89, 1988.
2. Cicchetti, DV. Guideline, criteria and rules of thumb for evaluating normal and standardized assessment instruments in psychology. Psychol Assess 6: 284-290, 1994.
3. Garhammer, J. Biomechanical profiles of Olympic weightlifters. Int J Sport Biomech 1: 122-130, 1985.
4. Gourgoulis, V, Aggeloussis, N, Antoniou, P, Christoforidis, C, Mavromatis, G, and Garas, A. Comparative 3-dimensional kinematic analysis of the snatch technique in elite male and female greek weightlifters. J Strength Cond Res 16: 359-366, 2002.
5. Gourgoulis, V, Aggelousis, N, Mavromatis, G, and Garas, A. Three-dimensional kinematic analysis of the snatch of elite Greek weightlifters. J Sports Sci 18: 643-652, 2000.
6. Hoover, DL, Carlson, KM, Christensen, BK, and Zebas, CJ. Biomechanics analysis of women weightlifters during the snatch. J Strength Cond Res 20: 627-633, 2006.
7. Isaka, T, Okada, J, and Funato, K. Kinematic analysis of the barbell during the snatch movement of elite Asian weight lifters. J Appl Biomech 12: 508-516, 1996.
8. Schilling, BK, Stone, MH, O'Bryant, HS, Fry, AC, Coglianese, RH, and Pierce, KC. Snatch technique of collegiate national level weightlifters. J Strength Cond Res 16: 551-555, 2002.
9. Stone, MH, O'Bryant, HS, Williams, FE, Johnson, RL, and Pierce, KC. Analysis of bar paths during the snatch in elite male weightlifters. Strength Cond 20: 30-38, 1998.
10. Thé, DJ and Ploutz-Snyder, L. Age, body mass, and gender as predictors of masters Olympic weightlifting performance. Med Sci Sports Exerc 35: 1216-1224, 2003.
11. Winchester, JB, Erickson, TM, Blaak, JB, and Mcbride, JM. Changes in bar-path kinematics and kinetics after power-clear training. J Strength Cond Res 19: 177-183, 2005.

barbell trajectory; phase diagram; relative barbell mass; biomechanics

Copyright © 2010 by the National Strength & Conditioning Association.