Avariety of training regimens are commonly used to improve power and work output of the athlete. Whether in sprinting, jumping, or throwing, the ability of the athlete to accelerate one's own body, an opponent, or an implement is crucial to sport performance. Plyometric exercises are a commonly used type of exercise in the development of power. Performing plyometric exercises makes use of the stretch-shortening cycle. In the stretchshortening cycle, a muscle is stretched directly before it is explosively contracted. This has been shown to allow the muscle to produce higher force and power outputs. One commonly performed plyometric exercise is the drop jump. In executing a drop jump, the athlete drops down from a height and directly on landing performs an explosive vertical jump. In the literature, drop jumps have been examined regarding various drop heights (5, 7, 13, 17) and differences in jump parameters regarding jump technique (1, 5, 11, 12). The interaction between jump technique and drop height has been reported only once in the literature to date (2). Others studies focus on the effects of drop jumps as part of a training program and their effect on performance (10). Lees and Fahmi reported (9) that drop jumps from 12 cm produced the highest vertical power output and jump height from starting jump heights of 12, 24, 36, 46, 58, and 68 cm. Other researchers (4) compared drop jumps from heights of 20, 40, and 60 cm and reported no difference in resultant jump height from the 3 starting heights. Yet other researchers (12) reported that countermovement jumps and drop jumps from 30 cm produced significantly higher jump heights, then squat jumps or drop jumps from 60 or 90 cm. A plausible reason for these discrepancies is that the authors are examining different jump techniques.
Results of plyometric training programs and their effect on performance are often inconclusive or inconsistent. Specificity seems to play an important role in designing a training program. For instance, when examining the effect of plyometrics on sprinting, we should differentiate between the acceleration phase and maintaining maximum velocity. Chelly and Denis (6) reported that during sprinting, power output showed a stronger correlation with acceleration, while leg stiffness (indicated by shorter contact times) showed a stronger correlation with maintaining speed. Arampatzis et al. (2) reported that drop jumps performed with shorter contact times also had higher leg stiffness values. The principle of specificity would lead us to the conclusion that different exercises would be ideal for acceleration than for maintenance of speed. Bobbert et al. (4) reported that during a bounce drop jump (a drop jump during which the athlete has a very short contact time), higher moments and power output were produced than during a countermovement drop jump (a drop jump in which the athlete has a longer contact time). This illustrates the fact that drop jumps from the same starting jump height can produce different jump parameter values.
Therefore, the purposes of this study are (a) to examine the relationship between contact time and force and power output and (b) to examine the interaction between starting jump height and jump execution or contact time.
The jump parameters that are considered important and measured in this study are mechanical power output and work performed on the center of mass (CM) and at the ankle and knee joints and moments at the knee and ankle joints as well as maximum vertical force values and vertical takeoff velocity.
Experimental Approach to the Problem
To examine the changes in important drop parameters, we instructed the subjects to perform drop jumps from heights of 20, 40, and 60 cm. The subjects performed the drop jumps with varying contact times. The subjects were then divided into groups based on their contact times, and the groups were compared with one another with respect to the chosen parameters.
A total of 15 athletes (decathletes, height: 1.83 ± 0.06 m, body mass: 78.94 ± 5.86 kg) participated in this study. The subjects performed drop jumps from 3 different heights (20, 40, and 60 cm). Informed consent was obtained from all subjects in accordance with the policy statement of the American College of Sports Medicine.
For control purposes, the subjects were instructed to keep their hands on their hips during the drop jumps. The instructions given to the subjects were (a) “jump as high as you can” and (b) “jump high a little faster (with relation to ground contact time) than your previous jump.” The ground contact time was measured and checked after every jump using a Kistler force plate. The jumps were performed at each height until the athlete could not achieve a shorter ground contact time. The first jump from each subject tended to have a contact time of over 200 ms. In the case that it was less than 200 ms, the subject was instructed to perform another jump but with a longer contact time. Five jumps per athlete per height were analyzed. The criteria for selecting the jumps to be analyzed were based on contact time. The first jump was supposed to have a contact time over 200 ms and the fifth jump the shortest contact time. The middle 3 should all have a contact time of less than 200 ms with a difference of 10 to 20 ms between jumps.
The human body was represented using a 15 segment 2-dimensional human body model (1). The masses and moments of inertia of the various segments were calculated using the data provided by Zatsiorsky and Selujanov (14). The ground reaction force was measured using a Kistler force plate (1,000 Hz). The movement of the athlete was captured using a high-speed (250 Hz) digital camera (redlake motionscope 250 C). The optical axis of the camera was approximately perpendicular to the plane of motion. To improve the quality of the video analysis, 6 reflective markers (radius 10 mm) were used to mark anatomical landmarks. The markers were fixed on the following body landmarks: the tip of the foot at the height of the metatarsals, the calcaneous, lateral maleolus, lateral epicondylus, trochanter major, and C7 vertebrae. These markers defined the position of the feet, lower legs, upper legs, and torso. The position of the head and arms in relation to the body remains constant in this model (Figure 1). The video was digitized using the Peak-Motus automatic tracking systems. The 2-dimensional coordinates were smoothed using a fourth-order lowpass Butterworth filter with a cutoff frequency of 15 Hz as described by Yu and Hay (13).
The vertical center of mass velocity during the support phase was calculated through integration of the vertical ground reaction force:
where Vz = vertical center of mass velocity, VzTD = vertical center of mass velocity at touchdown (this was calculated using the video data), Fz = vertical ground reaction force, and m = mass of the athlete, g = acceleration of gravity. The mechanical power was calculated by multiplying the vertical ground reaction force with the velocity of the center of mass:
where Pz = mechanical power of the ground reaction force. The joint moments and the mechanical power of the joint moments were calculated through inverse dynamics as follows (11):
where Mj = jth joint moment, MF = frictional moment, F = ground reaction force, rj = position vector between jth joint and the point of force application, rji = position vector between jth joint and center of mass of ith segment, Gi = force of gravity on the ith segment, pi = first derivative of the ith segment impulse, hi = first derivative of the ith segment angular momentum, n = number of segments.
where Pj = mechanical power of the jth joint moment and ωj = angular velocity of the jth joint.
The synchronization of the kinematic and dynamic data was achieved by starting the 3 measurement systems at the same time. All analyzed jumps were divided into 5 groups. Group 1 contained the jump from each subject with the longest contact time. Group 5 contained the jumps with the shortest contact times. Groups 2–4 contained the 3 remaining jumps from each subject, which were assigned to the given groups based on contact time with groups 2–4 have progressively longer contact times.
The differences among the groups were checked using a t-test for paired subjects. The level of significance was set at p ≤ 0.05. Pearson's correlation coefficients were calculated to examine the relationship between the different parameters.
All groups showed significantly (p < 0.05) different contact time values at each of the given starting heights (Table 1). Within the groups, there were no significant (p < 0.05) contact time differences with the exceptions of a couple heights in groups 1 and 5. The maximum vertical force values (VFMax) (Table 1) increased with both an increase in starting height and a decrease in contact time. The vertical velocity (Table 1) at takeoff was the highest in groups 1-for all starting heights. The takeoff velocities produced by groups 1-were all similar and significantly greater than those produced by groups 4 and 5 regardless of starting jump height.
Group 3 produced the highest maximum mechanical power output (CMPMax) values during the positive phase of the movement for all jump heights (Figure 2). Significant differences (p < 0.05) were seen between group 3 and all other groups at all starting heights. At all starting heights, the CMPMax value increased with decreasing contact time until a certain point at which the CMPMax started to decrease. Group 3 also produced the highest mean mechanical power output (CMPMean) (Figure 2) for all starting heights (p < 0.05). Within group 3, a starting height of 40 cm produced significantly higher values (p < 0.05) than from 20 or 60 cm. In contrast, the amount of work done by the CM (CMWPos) (Figure 2) during the upward phase of the movement was the highest in group 1 (p < 0.05) and the lowest in group 5 (p < 0.05) regardless of starting height. The differences between the groups were significant (p < 0.05) at all starting heights and between all groups except for groups 2 and 3. Groups 2 and 3 produced no significant differences at any of the starting heights.
The maximum mechanical power values produced at the knee joint (KPMax (Figure 3) were the lowest (p < 0.05) in group 5 regardless of the starting jump height. The values produced by group 4 were significantly less than those produced by groups 1–3 at all starting heights.
The highest values were found in groups 1 and 2. The maximum knee moment values (KMMax) were highest for each group when jumping from a starting height of 60 cm and the lowest for every group from a starting height of 20 cm (Figure 3). All the various starting heights produced significantly different KMMax values in all cases except for between 20 and 40 cm for groups 1 and 2. The lowest KMMax value was measured in group 5 from 20 cm. This value was significantly (p < 0.05) lower than the values from all other heights and groups. The work performed at the knee joint (KWPos) (Figure 3) was the highest in group 1 (p < 0.05) and the lowest in group 5 (p < 0.05) at all starting heights. The work performed by all groups was similar regardless of the starting height except in the following cases: group 3 values from 20 cm were significantly greater than those from 40 cm, and group 4 values from 20 cm were significantly greater than those from 40 and 60 cm.
The maximum power at the ankle joint (APMax) values showed a pattern that groups 3 and 4 produced higher values (Figure 3) than the other groups for each starting height. The various starting heights produced no significant difference within any of the groups. The maximum ankle moment (AMMax) showed a trend of increasing with decreasing contact time (Figure 3). No differences in maximum ankle moment (AMMax) were found between any of the groups with regard to starting height. The work performed at the ankle (AWPos) (Figure 3) joint was the lowest in group 5 (p < 0.05) for all starting jump heights. The highest values were found in groups 1–3, which at all starting heights produced similar values.
In reviewing the results, it can be seen that changes in jump contact time more often than not had a greater effect on jump parameters than starting jump height. This is in agreement with a conclusion drawn by Bobbert (3), who compared numerous drop jump training studies and found no patterns between an increase in performance and starting height, number of jumps per session, sessions per week, amount weeks trained, or whether the drop jumps were performed in combination with another form of training (i.e., weightlifting). The current authors agree with Bobbert (3) that the technique in these studies was an uncontrolled variable and that natural adjustments made by the subjects performing drop jumps under various conditions may have resulted in inconsistent results. We found that the CM power parameters, the AMMax, and the jump height were not affected much by starting height. On the contrary, the KMMax was markedly influenced be starting jump height with significant differences being measured between starting heights in most cases. In the present study, we measured contact times ranging from about 136 to 222 ms.
Group 3 (moderate contact time) produced the highest CMPMax and CMPMean values regardless of starting jump height. Groups 1 (jump as high as possible) and 5 (shortest contact time) produced the lowest CMPMax and CMPMean values regardless of starting height. In group 3, the highest values for CMPMax were produced with jumps from 40 and 60 cm (Figure 2), which were significantly (p < 0.05) higher values than those from 20 cm. The CMPMean values in group 3 for the 3 starting jump heights were not significantly different. For both parameters, group 3 (mean contact 161–167 ms) produced the highest values, and both slower and faster contact times produced lesser values. Voigt et al. (12) reported that drop jumps from 30 and 60 cm produced similar CMPMean values. This is in agreement with the results from this study that no difference was found in CMPMean values at 20, 40, or 60 cm due to starting jump height. In contrast to the CMPMax, CMPMean, and APMax values, the KPMax values were always the highest in groups 1 and 2 regardless of starting height and with shorter contact times tended to decrease. This means that in jumps in which there were relatively long contact times (179–222 ms), the knee contributed a larger percentage (˜45%) of the total power produced and that in jumps with slightly shorter contact times (160–167 ms), the contribution was reduced to ˜37%, and with even shorter contact times of 136–152 ms, the knee contributed only ˜28% of the total power. The APMax, on the other hand, produces ˜31–35% of the total energy regardless of contact time. For athletes who want to maximize both knee and ankle power values, it may be suggested to plan the training with consideration for the knee joint because the differences in the ankle joint values show a lot less variation with regard to changes in contact time and starting height.
The amount of positive work produced by the CM, the ankle, and the knee all seem to be influenced mainly by the contact time with the differences showing a general trend of more work being performed with longer contact times. This may be useful for planning preseason training where more work is likely to be performed and it is not yet critical that optimal power values and quickness of movement are present. As reported here, the APMax values were seldom affected by the starting jump height. It is possible that adjustments at the knee joint at different heights allow the ankle power values to remain similar.
As expected, the VFMax value increased steadily in conjunction with both increases in height and decreases in contact time (Table 1). The takeoff velocities in this study are comparable with those reported by Lees and Fahmi (9) for the lower heights but not for the higher heights. They reported a mean takeoff velocity values of 2.63 and 2.14 m·s-1 from starting heights of 12 and 58 cm, respectively. In the present study, takeoff velocities of ˜2.40 m·s-1 were measured from starting heights of 20, 40, and 60 cm.
These results indicate that the starting jump height should not be increased without consideration for contact time. This is an important finding for 2 reasons. First, in the literature there is traditionally more attention given to the starting height than the contact time. Several researchers (5, 9, 12) have published studies concerning starting jump height. Considering the results of this study, the search for an optimum starting jump height seems pointless without paying close attention to the contact times.
Second, in training situations and in general training publications, there is also much attention given to starting jump height. With regard to contact time, it is sometimes stated that the contact time should be minimized. An overemphasis on starting height in training situations is understandable because in training situations it is much easier to regulate starting height than it is to measure the contact time. In light of the results of this study, more attention should be given to contact time in training situations.
The difference in the power, work, and moment values illustrate the fact that the parameters trained during a drop jump vary, depending on how the drop jump is performed. Before designing a drop jump program as part of an athlete's training, the athlete/coach has to first decide what the goals of the training are. It appears that by manipulating the contact time of a drop jump, specific jump parameters can be targeted for training. An increase in starting height alone guarantees only an increase in FZMax. For CMPMax, CMPMean, and APMax values, a moderate contact time (161–166 ms for the athletes in this study) seems to be ideal. The KPMax, on the other hand, was maximized with longer relative contact times (179–222 ms in this study). A volleyball or basketball player, for instance, would tend to perform the stretch-shortening cycle in the form of a countermovement jump in which the push-off time is less restricted than, for instance, a long jump takeoff. The principle of specificity would indicate that during training jumps for maximum height less emphasis on contact time would be appropriate. It should also be noted that both volleyball and basketball players also perform maximum jumps directly after landing. The jumps would have a more limited contact time, and therefore training for these jumps would require jumps with somewhat shorter contact times. A sprinter, on the other hand, would want to train for maximum power output for the beginning of a race, when acceleration is the most important factor, but very short contact times for the speed maintenance phase, when the contact times are more restricted.
These findings also have an effect on current diagnostic activities and experimental procedures. For example, if a treatment is given to a group of subjects with a pre- and post-vertical jump test, an improved posttest vertical jump performance without consideration of jump technique is not necessarily an indication that the subject actually improved his or her jumping ability. It is plausible that the subject unknowingly altered the contact time and through this alteration produced a different jump. A suggested direction for further research is to examine knee and ankle power and contact times during various sports. This would aid in designing plyometric programs suited for the needs of specific sports.
1. Arampatzis, A., and G.-P. Brüggemann. A mathematical high bar-human body model for analyzing and interpreting mechanical-energetic processes on the high bar. J. Biomech.
2. Arampatzis, A., F. Schade, M. Walsh, and G.-P. Bruggemann. Influence of leg stiffness and its effect on myodynamic jumping performance. J. Electromyogr. Kinesiol.
3. Bobbert, M.F. Drop jumping as a training method for jumping ability. Sports Med.
4. Bobbert, M.F., P.A. Huijing, and G.J. Van Ingen Schenau. Drop jumping. I. The influence of jumping technique on the biomechanics of jumping. Med. Sci. Sports Exerc.
5. Bobbert, M.F., P.A. Huijing, and G.J. van Ingen Schenau. Drop jumping. II. The influence of dropping height on the biomechanics of drop jumping. Med. Sci. Sports Exerc.
6. Chelly, S.M., and C. Denis. Leg power and hopping stiffness: relationship with sprint running performance. Med. Sci. Sports Exerc.
7. Clutch, D., M. Wilton, C. McGown, and G.R. Bryce. The effect of drop jumps and weight training on leg strength and vertical jump. Res. Q. Exerc. Sport
8. Hof, A.L. An explicit expression for the moment in multibody systems. J. Biomech.
9. Lees, A., and E. Fahmi. Optimal drop heights for plyometric training. Ergonomics
10. Rimmer, E., and G. Sleivert. Effects of a plyometrics intervention program on sprint performance. J. Strength Cond. Res.
11. Schmidtbleicher, D. Klassifizierung der Trainingsmethoden im Krafttraining. Leichtathletik
12. Voigt, M., E.B. Simonsen, P. Dyhre-Poulsen, and K. Klausen. Mechanical and muscular factors influencing the performance in maximal vertical jumping after different prestretch loads. J. Biomech.
13. Yu, B., and J.G. Hay. Angular momentum and performance in the triple jump: A cross sectional analysis. J. Appl. Biomech.
14. Zatsiorsky, V.M., and V.N. Selujanov. The mass and inertia characteristics of the main segments of the human body. In: Biomechanics VIII-B.
H. Matsui and K. Kabayashi (eds.). Champaign, IL: Human Kinetics, 1983. pp. 1152-1159.