Kinematics Analysis of Lunge Fencing Using Stereophotogrametry

Authors: 1M. Gholipour, A. Tabrizi and 2F.
1Farahmand department of Physical Education, Sharif University of Technology,Tehran, Iran 2Mechanics Faculty, Sharif University of Technology,Tehran, Iran

Abstract: The purpose of the present study was to cinematically analyze fencing lunge in two groups of elite and novice athletes. Eight right-handed fencers (foil-male) were selected from Sharif University’s fencing team (novice group, mean 21.5±1.3 years) and Iranian national fencing team (Elite group, mean 24±2.5 years) and were tested in two different days following installing of markers over subjects’ joints. They performed fencing lunge, facing three high speed Kinemetrix cameras. The results showed that the elite group had a higher mean lunge length than the novice group (1.17±0.17m vs. with 1.02±0.1 m). Studying the affecting factors, it was noticed that the elite group inclined their bodies (0.117±0.04 m) more towards their front legs than novice group (0.051±0.30) in on-guard position. Investigation of the joints angular motion revealed that the initial knee flexion in elite group (mean 20±12 deg) was less than novice group (mean 38±15 deg) but their following extension in the middle phase of motion (mean 51±9 deg) was considerably larger (mean 18±8 deg). The angular motion of the hip joints of the two groups was negligible except for the final lunge position. In contrast to the popular belief, that the motion of armed hand and blade precedes the leg motion, it was observed that the armed hand and leg moved simultaneously in elite group. Our results emphasize the necessity of strengthening the quadriceps and hamstring muscles, in addition to accurate observation of the lower limbs motion pattern, to achieve a longer lunge.

Key words: Fencing » Lunge » Stereophotogrametry » Kinematics analysis


Most of the previous investigations on fencing have studied the physical and morphological characteristics of elite fencers or the biochemical and nutritional factors that affect the athletes’ performance [1]. These statistical studies are often much difficult to provide general conclusions, considering the variety of the affecting factors e.g., length, weight, muscular strength and flexibility of the athlete [2]. However, some researchers have successfully correlated the physical characteristics of fencers with their success in competitions and suggested some indexes for evaluating their conditions [3-5]. A literature review shows that there has been a limited number biomechanical investigations concerning fencing. Lunge movement has been one of the few subjects mostly studied by biomechanics researches. Klauck and Hassan (1998) studied the kinematics of fencing movements in a group of elite female fencers using an electro-optical system and reported that even high-level champions do not follow standard patterns; each fencer follows his own pattern to perform the required movements [6]. Another research showed that the motions of the hip joint and the armed hand of fencers started simultaneously [7]. Legnani et al. (1999) installed markers on the body and Foil of right-handed fencers with different skill levels and studied the displacement, velocity and acceleration of different points on their body [8]. Zhang et al. (1999) studied the lunge technique of 4 elite female fencers during competition using photogrammetry and measured the kinematic variables such as the lunge length, reaction time, horizontal speed of the center of the gravity and the time to reach the target.

In spite to the popular belief indicating that the motion of the weapon precedes the body motion and the body incline precedes due to the blade motion, it was found that the fencer’s center of the gravity and the sword move simultaneously [9].
Dennerlein and Galea (2000) investigated the relationship between the strike force and kinematical variables  such as velocity  and lunge  stride using photogrametry [10] and found that there was a direct relationship between impact velocity and the imparted force, with a higher force imparted partway through a full lunge. Stewart and Kopetka (2005) investigated the kinematical factors that affect the lunge speed through photogrammetric analysis of 15 elite Eppe fencers. They reported that the lunge speed is depend on the maximum angular velocity of the sword arm elbow and knees, however, a more important factor is the time interval in which they can be reached [11].

As the above review shows, studying the kinematical characteristics of the lunge skill has been one of the most attractive topics of research in fencing, considering the importance of the high lunge speed due to the harmonic motion of all body parts. This has been expected to provide quantitative criteria and optimum patterns by studying and extracting kinematics of successful fencer’s motion in which can be used to train beginners. Considering the above fact, the present study was conducted, as a complementary to the previous investigations, to extract and compare the kinematical variables of two groups of elite and novice fencers, including the joint angles, stride length and traveling distance and speed of the lunge motion, using stereophotogrametry method.

In order to investigate and analyze a sport movement, it is necessary to understand its nature and goal and to identify the factors that affect its performance. The lunge movement in fencing includes 3 main successive steps through which the weapon should have a steady motion towards the opponent. First, the tip is put in line the armed hand is extended and sword is pointed to the target (Fig. 1a). Second, through a fast motion, immediately on having the arm extended, the front leg is kicked forward and also pulling the pelvis and the body center of the gravity forward. Finally, in the third step, once the front leg has been kicked, the back leg straightened quickly. All of the above motions should be executed in one smooth, seamless and harmonic set of actions so that at the end of motion the back leg is completely extended, the front shank and sword extended arm are nearly perpendicular and parallel to the floor respectively (Fig. 1b) [12]. Considering the fact that both the upper and lower limbs of the armed and non-armed sides contribute to the lunge movement, an exact analysis of the movement needs the position of all limbs (in both side of the body) to be recorded and studied in detail. On the other hand, reaching the maximum speed and traveling distance are the most important characteristics of the lunge motion. This needs observation of some kinematics points during lunge execution which in fact distinguishes elites from novice fencers.

Subjects included 4 right-handed Foil fencers of Sharif University of Technology (the novice group) and 4 right-handed fencers from the national team of the Islamic Republic of Iran (the elite group). Table 1 represents the general characteristics of the subjects. The two groups were tested in two separate days and their movements were recorded and analyzed. First, each participant devoted some time to warm up and performed multiple lunge movements to get acquainted with the testing environment. In order to record the movement, the anatomical landmarks including the metatarsophalangeal joint (MTP joint), medial malleolus (ankle joint), medial epicondyle of femur (knee joint), anterior superior iliac spine (pelvis), acromion process of scapula (shoulder joint), medial epicondyle of humerous (elbow joint), ulnar styloid process (wrist joint) in both sides of the body were used to indicate foot, shank, thigh, pelvis, trunk, arm and forearm segments respectively. After identifying the anatomic positions of landmarks, light reflecting markers(2.5 cm diameter) were attached to them by two-sided pastes (Fig. 1).

Table 1. Characteristics of the subjects Athletes

Novice Elite
1 2 3 4 Mean±SD 1 2 3 4 Mean±SD
Age (years) 21 20 22 23 21.5±1.3 26 27 23 20 24.0±2.5
Length (cm) 180 177 180 180 179.3±1.1 191 176 179 180 181.5±4.8
Weight (kg) 78 64 63 76 70.3±6.8 80 76 75 65 74.0±4.5


The positions of markers installed on the athlete's body

Fig. 1. The positions of markers installed on the athlete’s body

Each participant took part in 10 tests. In each test, the subject first took the guard position in a specified place in front of the cameras and got ready to lunge. Then, with the order of the tester, the subject performed the lunge technique as fast and strong as possible. The motion was recorded using 3 high-speed kinemetrix cameras (50 fps). The general pattern of motion and the path of markers as recorded by one of the cameras are represented in Fig. 2. In the next step, the data of the 3 cameras were integrated using motion analysis software to obtain the 3D coordinates of each marker during the movement execution. Figure 3 shows the paths of markers in the sagittal plane during execution of lunge technique by one of the elite fencers.

The paths of markers installed on the body joints of a novice fencer form start to end of the lunge movement as recorded by one of the cameras

Fig. 2. The paths of markers installed on the body joints of a novice fencer form start to end of the lunge movement as recorded by one of the cameras


The computed sagittal plane paths of markers installed on the body of an elite fencer during performing the lunge technique

Fig. 3. The computed sagittal plane paths of markers installed on the body of an elite fencer during performing the lunge technique

In the next step, the data recorded for each subject was investigated and the test with the longest lunge was chosen. To measure the length of lunge, the coordinates of the markers installed on the front ankle at the beginning of the motion (on guard position) and the end of lunge were compared. To evaluate the lunge speed, the traveling distance and time for the marker installed on the iliac spine of the armed side (H-curve in Fig. 3) were measured from beginning to the end of motion. Moreover, the ASIS displacement velocity obtained using the coordinates of the attached markers in on-guard position and at the end of motion, was used to find the displacement of the body center of the gravity. To evaluate the level of athlete’s reliance on front leg before starting the movement (on-guard position), the x coordinate of the markers installed on front ankle and ASIS were compared. Finally, the joints angles of the front leg and armed hand were computed using the coordination data of the relevant markers and some mathematical calculations using EXCEL. Considering the limited number of subjects, the statistical comparison of the results of the two groups was not conducted and only the means and standard deviations of variables were studied and compared. Moreover, considering the limited number of cameras, the motion of some effective body parts such as the hip joint of the front leg (armed side) was not analyzed. Also, due to some software limitations, in spite of the availability of the 3D coordination of markers, this study was limited to 2D motion analysis in sagittal plane. Finally, to reduce the displacement of markers on the subject’s body, which is a major concern in all similar photogrammetric investigations, the markers were installing directly on the subject’s skin.



The means and standard deviations of elite and novice group’s lunge length were found to be 1.17±0.17 and 1.02±0.1 meters, respectively. Evaluation of the lunge speed using the traveling distance and time of markers installed on the iliac spine of armed sides showed that the elite and novice group moved 1.14±0.18 and 1.06±0.11 meters, respectively, in 1.82±0.67 and 1.46±0.27 seconds. The motion starting times of the armed hand and front foot and thereby lunge were compared by analyzing the motion starting times of the elbow and knee joints. Results indicated that the hand motion preceded the foot motion by 0.13±0.15 seconds in novice group and 0.07±0.05 seconds in elite group. Also, by calculating the difference between the means of the positions of ASIS and front ankle markers, as an index for evaluating the anterior inclination of the center of gravity and consequently the trunk, the inclination values in novice and elite groups were obtained to be 0.05±0.02 and 0.12±0.03 meters, respectively.

Table 2. The results of the measurements conducted during the tests

 Measurement Novice group Elite Group
Mean±SD Mean±SD
Lunge length (m) 1.02±0.10 1.17±0.17
Traveling distance of ASIS (m) 1.06±0.11 1.14±0.18
Traveling time of ASIS (sec) 1.46±0.27 1.82±0.67
Difference of motion starting time of hand and foot (sec) 0.13±0.15 0.07±0.05
Trunk anterior inclination (m) 0.05±0.02 0.12±0.03 


Figure 4 and 5 represent the knee and hip joints angular motions in elite and novice groups. To evaluate the knee flexion (the angle between ankle, knee and ASIS markers in armed side) and hip joint flexion (the angle between knee, ASIS and shoulder markers in armed side), their angle in on-guard position was considered to be zero and the following changes during motion were computed. The general pattern of knee and hip joints flexion during long movement were nearly similar in the two groups. All participants increased their knee flexion at the beginning of motion and then threw their legs to the front while extending the knees (the angles between H, J and L markers in Fig. 3). The values of the flexion angles, however, were different in the two groups. The mean flexion angle of the knee in the first one-third interval of motion in elite and novice groups were respectively

Knee joint angular motions in the two groups
Fig. 4. Knee joint angular motions in the two groups


Hip joint angular motions in the two groups

Fig. 5. Hip joint angular motions in the two groups


20±12 and 38±15 degrees and the means of its subsequent extension in the elite and novice groups were respectively 51±9 and 18±8 degrees. Similarly, the hip extension in the first one-third interval of the motion was 2±2 degrees for elites and 4±5 degrees for novices and means of its subsequent flexion in medial phase of motion were 37±3 degrees for elites and 40±7 degrees for novices. The means of the hip joint flexion in the final phase of motion and obtaining the lunge position were 53±11 and 40±7 degrees for elite and novice groups, respectively.



Lunging is one of the most fundamental movements of fencing and other rocket sports. It has been shown that the right and fast execution of lunging and its subsequent movements (return to the initial position or change of rout) has a crucial role in athletes’ success [2]. However, there are a limited number of investigations concerning the kinematical pattern of lunge movement. The most comprehensive research in this field is the work of Stewart and Kopetka (2005) who studied the kinematics of lunge in 15 Eppe fencers using photogrametry with 50 Hz cameras. In their research, fencers were requested to perform fencing lunges as fast as possible in response to a light stimulus. Results of recording the motion of 24 markers were analyzed based on the angular velocities of joints and it was shown that the maximum angular velocity of the armed elbow is the only variable is dependent on the overall speed of lunge among the 6 independent and dependent variables studied [11]. The present study is similar to this research in terms of the test method. However, it provides new useful information through testing two groups of elite and novice fencers and analyzing the results based on the joints angles, rather than the joints angular velocities.

The results of the present research showed that in spite of the almost similar height average, the elite fencers were able to perform longer lunges. It seems that this ability is related to the more anteriorly inclined center of gravity and consequently the trunk of this group, just before throwing the lunge. As mentioned previously, the elite group inclined their center of gravity by about 6 cm more anterior than the novice group. With the body inclined anteriorly, the load and weight on the back leg is decreased and so it can take a more effective role in throwing the body to the front by extending the knee. This is an important characteristic which should be considered in training novice fencers.

Although the ability of a fencer to perform a long lunge is an effective factor for his or her success, it would be much more effective when performed with a high speed providing less chance for opponent’s reaction. The results of the present study indicated that, in spite of having a longer lunge, the elite group had a slower motion than the novice group; the novice fencers traveled 1.06 m in 1.46 seconds (mean velocity 0.76 m/s) and elite fencers traveled 1.14 m in 1.82 seconds (mean velocity 0.68 m/s). Although the difference observed was small, it emphasizes on the necessity of paying more attention to high speed motion in fencing exercises.

It is a common belief among fencer trainers that the motion of armed hand should precede the front leg’s motion during lunge. However, the results of the present study indicated that the elite fencers moved their armed hands later than their feet, in comparison with the novice group. In fact, in spite of the novice group, the elite group moved their elbow joints and consequently their upper limbs with their knee joint simultaneously. This finding is consistent with the results reported by Zhang et al. (1999) and Klauck and Hassan (1998) and rejects the mentioned belief [7 and 9].

The results of this study concerning the knee and hip joints angular motion during lunge movement indicated that both the elite and novice groups followed almost a same pattern. However, detailed and quantitative investigation showed that the initial knee flexion in elites (mean 20°) was smaller than novices (mean 38°) and the knee extension in the medial phase was much larger for elites (mean 51°) than the novices (mean 18°). These differences indicated the importance of knee joint motion in throwing the leg to the front to obtain a longer lunge; a fact that has been neglected in previous studies. The hip joint motion in the two groups was nearly the same except for at the final stage of lunge movement. It seems that this higher hip flexion in the elite group (mean 53°) in comparison with the novice group (mean 40°) is correlated with their higher knee extension in this stage and is conducted to decrease the height of the center of gravity and provide more stability. The lower hip flexion and knee extension of the novice group might be due to their lower muscle flexibility and strength. The above results, in addition to emphasizing on the importance of the kinematical pattern of the lower limb joints, highlights the importance of the muscles strengthening, especially the quadriceps muscles for knee joint extension and the hamstring muscles for hip joint flexion [3, 4].

Comparison of the participants kinematics within each group, shows that there was a relatively smaller difference in the joints angular motion of the elites than that of the novices; this was more significant for the knee joint (Fig. 5). However, the considerable differences observed in the joints angular motions (especially for the hip flexion in the final stage of the movement) and their occurrence times for the members of the elite group indicates that each participant follows its own detail pattern to perform the technique, in spite of the overall similar pattern. This finding is consistent with the results reported by Hassan and Klauck (1998) [6].



  1. Harmenberg, J. and R. Ceci, 1991. Fencing: Biomedical and psychological factors.
  2. Cronin, J.P., J. Mcnair and N. Marshall, 2003. Lunge2 performance and its determinants. J. Sports Sci., 21(1): 49-57.
  3. Sapega, A.A., J. Minkoff, M. Valsamis and J.A. Nicholas, 1984. Musculoskeletal performance testing and profiling of elite competivtive fencers. Clinics Sports Medicine, 3(1): 231-244.
  4. Sapega, A.A., J. Minkoff, J.A. Nicholas and M. Valsamis, 1978. Sport-specific performance factor profiling: fencing as a prototype. American. J. Sports Medicine. 6(5): 232-235.
  5. Stewart, K.J., A.R. Peredo and C.M. Williams, 1977. Physiological and morphological factors associated with successful fencing performance. J. Hum Ergol (Tokyo), 6 (1): 53-60.
  6. Klauck, J. and S.E.A. Hassan, 1998. Lower and upper extremity coordination parameters during the fencing lunge. Abstracts from ISBS symposium XVI.
  7. Hassan, S.E.A. and J. Klauck, 1998. Kinematics of lower and upper extremities motions during the fencing lunge: results and training implications. Abstracts from ISBS symposium XVI. www.isbs98. uni
  8. Legnani, G., B. Zappa G. Roi and M. Galli, 1999. Dynamic simulation of fencing hits. VII international symposium on computer simulation in biomachanics, Calgary, 5-7: 156-160.
  9. Zhang, B.M., D.P.K. Chu and Y. Hong, 1999. Biomechanical analysis of the lunge technique in elite female fencers. Abstracts from ISBS symposium XVII.
  10. Galea, A. and J. Dennerlein, 2000. Schlager fencing biomechanics: determinates of impact force. Annual conference of the American society of biomechanics.
  11. Stewart, S.I. and B. Kopetka, 2005. The kinematic determinants of speed in the fencing lunge. J. Sports Sci., 02 (Part I. Biomechanics)
  12. Rudy, V.,  1997. Big book of fencing Publisher: Rudy Volkmann; Spiral edition, ISBN 0-9668038-0-9 pp: 52-58.

1950 Santelli: Foil Fencing Fundamentals

1950. Silent. Foil Fencing Fundamentals. Instructor: Giorgio Santelli, Coach of the US Olympic Fencing Team. Student: Robert Nielsen, 1950 Collegiate Foil Champion. Video provided by Giorgio’s son, John Santelli, who wrote, “However, gratified to see such interest, I just made a DVD of a professionally-made silent lesson video from 1950… ” He later added, “Glad you got ’em ok. Feel free to add them to your fine YouTube collection (which I found by happenstance)… way more folk will get to see ’em in your collection than if I uploaded them! Best regards, John.”