Introduction
Sprint training usually includes speed, speed endurance, and strength training. Strength training of sprinters comprises hypertrophic, maximum-, and speed-strength training. By hypertrophic training (60–80% 1RM [repetition maximum]) sprinters are trying to increase the cross-sectional area of muscles (especially that of fast muscle fibers) and thereby increasing maximum strength. By maximum strength training (90–100% 1RM), athletes are trying to improve the function of their nervous system (e.g., rapid and increased recruitment of motor units, increased firing rate of motor units, and improved synchronization of motor units) (10,22,35) and to activate the largest fast twitch motor units (11,34), although it might be also possible with rapid ballistic muscle actions (34).
In speed-strength training, loads are lighter than in maximum strength training (and usually lighter than in hypertrophic training) and thereby movement speeds are faster and closer to the movement speeds of sprint running. Examples of speed-strength training include ballistic exercises (e.g., squat jump), plyometric exercises (jumps), and sprint specific exercises (e.g., sled pulling). It is thought that the effects of strength training are movement and velocity specific so that adequate employment of plyometric and sprint-specific exercises is important in sprint training (10,38).
However, in training studies where adaptations to different kinds of training programs (different training velocities) have been studied, the results have been versatile according to the velocity specificity in training adaptations. Training and velocity-specific adaptations have been found after normal, resisted and supramaximal sprint training, and after isokinetic training with different velocities (12,23,32,40). At the same time, in few other studies (3,9,17), the adaptations to high- and low-velocity training did not show signs of velocity specificity because there were no clear differences in sprint running adaptations between the groups. However, the possible reasons for this could be, for example, that high- and low-velocity training were combined with sprint training and that differences between training velocities between the groups may have not been large enough.
Although velocity specificity is thought to be important in training, it can be hypothesized that the neuromuscular system would adapt to that kind of training, and after a while, no further adaptation would most likely occur. In this kind of scenario, it could be advisable to add some exercises wherein force production and velocity of the movement would be higher than in a certain sport performance. This kind of training could lead to improvements in force production and velocity during a movement (positive changes on the force-velocity curve) and to an improvement in a certain sport performance (7,25). Furthermore, there is some evidence (32) that training adaptations (in force production) in higher angular velocities transfer well to lower angular velocities.
Despite the common use of speed-strength and strength exercises in training of sprinters, there is a lack of knowledge of biomechanical characteristics of these exercises (e.g., force production, angular velocities, electromyographic [EMG] activity of the essential muscles with regard to the sprint start) as compared with sprinting (24,26,30). However, it would be important to know how different exercises used in the training of sprinters relate to the sprint start to better understand how different exercises influence on the sprint start performance. Ultimately, this information could probably be used in more sensible planning of the training of the sprint start, based on earlier studies of training adaptations (14,15,33), specificity (2,7,12,25,27,32), and EMG activity during the sprint start (29).
Hence, the purpose of this study was to compare kinetics, kinematics, and muscle activity between the sprint start, sled pulling, and selected squat-type exercises and, thus, to acquire more information about the relations between these exercises (force production, angular velocities [contraction velocities], and EMG activity). In addition, the aim was to study how different exercises are related to a good performance in the sprint start and, thus, correlations between the sprint start performance and the studied exercises were examined. Based on the earlier biomechanical studies concerning the studied exercises (5,8,16,21,28,36), it was hypothesized that (a) the force production is higher during the 1/2-squat as compared with that during the block phase of the sprint start, (b) the angular velocities of the hip and knee joints are higher during the countermovement jump (CMJ) as compared with that during the block phase of the sprint start, and (c) that CMJs with moderately light loads correlate better with the sprint start performance (10 m) than the 1/2-squat with heavy loads.
Methods
Experimental Approach to the Problem
The study was designed to compare the sprint start to different exercises used typically in sprint training and to examine how these exercises correlate with the sprint start performance (10-m time). These exercises were sled pulling, CMJ, and 1/2-squat performed with various loads. To investigate this question, 9 track and field athletes performed the above-mentioned exercises, and different force time, EMG, and kinematic variables were measured and analyzed. The main reason to investigate force-time variables was to find out if force production was higher in the studied exercises and, hence, if they could have potential to induce positive changes on the force-velocity curve if used in training. The EMG variables were studied to find out how the essential muscles concerning the sprint start were activated in the studied exercises. The kinematic variables were studied to discover how velocity and movement specific to the studied exercises were as compared with the block start and if the exercises could have potential to induce positive changes on the force-velocity curve. General linear model repeated measures analysis of variance was used to investigate the statistical differences between exercises, and Spearman's rho was used to find out how studied exercises correlate with the sprint start performance. This design enabled us to evaluate what kind of similarities and differences (in various kinetic, kinematic, and EMG variables) exist between the sprint start and studied exercises and which exercises (and qualities) could be important for a good sprint start performance.
Subjects
Nine male athletes (4 sprinters, 3 decathlonists, 1 long jumper, and 1 triple jumper; age = 24.9 ± 3.9 years; height = 180.4 ± 5.9 cm; weight = 80.3 ± 7.5 kg; 100-m record = 11.35 ± 0.29 seconds; track and field training years = 11.8 ± 3.1 years) participated as subjects in this study. All the subjects were familiar with the exercises performed in this study and provided informed consent and health inquiry before the study. Approval for the study was obtained from the Ethics Committee of the Central Finland Health Care District.
Procedures
At the beginning of the testing session, each subject filled out an informed consent and health inquiry form. Then the weight and height of the subject were measured, and the locations of the surface EMG electrodes were defined (Surface ElectroMyoGraphy for the Non-Invasive Assessment of Muscles). Thereafter, each subject performed his individual warm-up that lasted 30–40 minutes. After the warm-up, electrode locations were slightly treated with abrasive paper and alcohol to reduce the impedance (<10 kΩ). Electrodes (Blue sensor N; Ambu A/S, Ballerup, Denmark) were attached on the leg that was in front of the starting blocks. Electromyographic activity was measured with bipolar configuration (interelectrode distance ∼20 mm) from the gluteus maximus (GM), biceps femoris (BF), vastus lateralis (VL), and gastrocnemius medialis (Ga). Tape and bandages were used in electrode attachments to ensure contact between electrodes and skin during the measurements. Each subject then tested starting blocks (Nordic Sport, Skellefteå, Sweden) and after attaching the markers (for 2D movement analysis) he was ready to start the measurements. The marker locations were the head of the fifth metatarsal, calcaneus, lateral malleolus, lateral epicondyle of the femur, greater trochanter, and middle portion of the neck. In CMJs with a load and in the 1/2-squat with a load of 70% 1/2-squat 1RM, the marker of the neck was replaced by the marker in the head of the weightlifting bar.
The measurements included the following exercises: block start (10 m; Blocks), sled pulling with the load of 10 and 20% of body mass (BM; 10 m) (Sled10 and Sled20), 1/2-squat 1RM in the Smith machine (SMAX) (Kraftwerk, Hot Milling Ltd., Tuusula, Finland), CMJ with the load of 10 and 20% of 1RM in the 1/2-squat (CMJ10 and CMJ20), and 1/2-squat with the load of 70% of 1RM in 1/2-squat (S70). The depth in the 1/2-squat exercises (knee angle 90°) was controlled by a rubber band and visual control. All the exercises, except 1RM 1/2-squat, were performed twice. If the subject performed the 2RM squat, 1RM was calculated (1). The exercises were performed in the above-mentioned order, and recovery times between repetitions and exercises were 2–5 minutes to minimize accumulation of fatigue. The exception was 1RM 1/2-squat testing, which included a specific warm-up (1/2-squatting with increasing weights), and the recovery after 1/2-squat testing was at least 10–15 minutes before performing the next exercise (CMJ).
The measurements were executed during the end of the November and the beginning of the December (∼1–2 months before the beginning of the indoor season). To ensure a good arousal level during the testing sessions, the starting time of the measurements was between 10 AM and 2.30 PM, and the subjects were instructed to keep the previous day as a resting day or as a light training day.
Equipment
In all the exercises, ground reaction forces (GRFs) were measured by the strain gauge sensor force plates (Raute PLC, Lahti, Finland; sampling rate 1,000 Hz). Electromyographic activity was measured by the wireless EMG system (Telemyo DTS, Noraxon USA Inc., Scottsdale, AR, USA; sampling rate: 1,500 Hz, gain: 1,000-fold, band-pass filter: 10–500 Hz). All the exercises (except the 1/2-squat 1RM) were also recorded by high-speed digital video cameras (Fastec Imaging, Fastec InLine, Fastec Imaging Corporation, San Diego, CA, USA; frequency: 250 frames per second, shutter speed: 1/500, resolution: 640 × 480). High-speed cameras were calibrated before every subject with the calibration frames (200.2 × 200.2 cm [Blocks, Sled10, Sled20] and 84.2 × 200.2 cm [CMJs and S70]). The data from the force plates and EMG system passed through the analog-to-digital converter (CED Power 1401, Cambridge Electronic Design Ltd., Cambridge, United Kingdom) to the computer and were recorded and analyzed with the Signal-software (Signal 4.04, Cambridge Electronic Design Ltd.). High-speed digital video camera data were recorded by another computer with the FIMS software (FIMS 3.0.4, Fastec Imaging Inc., San Diego, CA) and analyzed with the Vicon Motus-software (Vicon Motus 9.2, Vicon Motion Systems Inc., Los Angeles, CA, USA). Before analyzing with the Vicon Motus, data were converted with the MiDAS Player 2.2.0.8 (Xcitex Inc., Cambridge, MA, USA). The measurements were started by manual triggering.
Blocks, Sled10, and Sled20 were performed on the strain gauge force plates that were embedded in the ground. The force plates covered the distance of 10 m and were surfaced with normal synthetic running track. Blocks were performed from the starting blocks (Figure 1A) and Sled10 and Sled20 were performed from the “3-point starting position” (Figure 1B). All the sprints were performed with spiked shoes. Sled pullings were performed with a custom-made sled (mass, 4.9 kg; Figure 1B). The sled was attached to the waist of the subject with a rope (17.4 m long) and a powerlifting belt. The force plates measured GRFs during the block phase (the phase when force is produced toward the starting blocks) and at least 6 following steps (used in the calculation of step frequency). The GRFs were measured separately from both the legs. Photocells were used in timing.
;)
Figure 1: A) Starting blocks and their placement on the force plates. B) Three-point starting position and custom-made sled.
The best trial in each exercise for each subject was analyzed. A clear increase in the horizontal force (anterior-posterior) was used as the mark for the beginning of the performance, and the signal from the photocells at the 10-m mark ended the performance. In most analyses, only the “block phase” (phase of force production toward starting blocks [or the corresponding phase in Sled10 and Sled20]) was used. The only exception was step frequency, which was presented as a mean value of the first 6 steps. Resultant peak forces (RPFs) were calculated at the moment of vertical peak forces. Resultant impulses (Res IMP) and angle of force productions (angle of FP) were calculated from vertical (net) impulses (Ver IMP) and horizontal impulses (Hor IMP). The takeoff velocity in the block phase was computed from Res IMP by dividing it with the BM of the subject measured on the force plate.
In the EMG analyses, the EMG signal was first rectified and then integrated to obtain integrated EMG (IEMG) values for the block phases. To obtain averaged EMG (aEMG) values, IEMG values were divided by the corresponding times. The EMG values were normalized individually to the EMG values of the concentric phase of CMJ20. Because of some technical problems in the measurements, the EMG values of the Ga were not analyzed in this study. In the kinematic analyses, the beginning of the block phase (Blocks, Sled10, and Sled20) was defined as a moment when the hip joint started to extend (threshold angle 20°·s−1). The end of the block phase was defined as the moment of takeoff (defined visually). In this study, 180° refers to a full extension of a joint. Force and EMG data were analyzed manually with the Signal software (Signal 4.04). Kinematic data were analyzed mostly automatically with the Vicon Motus software (Vicon Motus 9.2).
Reproducibility of some dependent variables related to the sprint start has been studied in some earlier studies, which are as follows: Linear correlation coefficients and coefficient of variation (CV) for horizontal average force (r = 0.96 and CV = 2.9%), vertical average force (r = 0.94 and CV = 4.2%), and force production time (FPT; r = 0.93 and CV = 3.6%) (28). Intraclass test-retest reliabilities for sprint times 5–20 m (r ≥ 0.90) (7).
Countermovement jumps and 1/2-squats were performed on the force plates embedded in the ground. Countermovement jumps with a load (CMJ10 and CMJ20) and S70 were performed with a weightlifting bar (20 kg) placed on the shoulders. Additional weight plates were used in CMJ10 (when necessary), CMJ20, and S70. The SMAX was performed using the Smith machine. Only the concentric phases of the exercises were used in the comparison with the block start.
The best trial in each exercise for each subject was analyzed. The beginning of the concentric phase was computed from the force-time signal, and the takeoff velocity in CMJs was computed from net concentric impulse by dividing it with the system mass (BM + mass of the load) measured on the force plate. Otherwise, the force-time, EMG, and kinematic analyses were executed in a similar way as the corresponding analyses of the block start and sled pulling.
Reproducibility of some dependent variables related to the CMJs has been studied in some earlier studies and are as follows: Intraclass test-retest reliabilities for force-time curve and velocity-time curve during the CMJ: r ≥ 0.90 and r ≥ 0.89, respectively (6). The corresponding values for aEMG during the CMJ, and minimum and maximum angles during CMJs (with various loads: 0, 20, 40, 60, and 80% of 1RM): r ≥ 0.80 and r ≥ 0.92, respectively (7).
Statistical Analyses
Standard statistical methods were used for the calculation of means and SD. General linear model repeated measures analysis of variance (PASW Statistics 18 software) was used to compute statistically significant differences between the block start and the other studied exercises. In the few cases, where the values were not normally distributed, the Wilcoxon signed ranks test was used to examine the statistically significant differences. Spearman's rho was used to calculate the correlation coefficients between performance time in block start (10 m) and other variables. An alpha level of p ≤ 0.05 was chosen as the criterion for significance.
Results
Average 1/2-squat 1RM was 185.5 kg (range, 167–227 kg) and average 1RM/BM was 2.32 (range, 1.91–3.01). The average loads used in CMJ10, CMJ20, and S70 were 11.0% (20.3 kg), 20.6% (38.1 kg), and 70.2% (130.3 kg) of the SMAX, respectively. The average loads used in Sled10 and Sled20 were 9.9% (7.9 kg) and 20.1% (16.1 kg) of the BM, respectively.
Force-Time Variables
Performance times (10 m), FPTs, and RES IMP were significantly (p ≤ 0.05) greater and step frequency (over 6 first steps) significantly smaller in Sled10 and Sled20 than in Blocks. The FPTs in the concentric phase during CMJ and CMJ10 were significantly shorter, and FPTs in 1/2-squats were significantly longer than in the block phase of Blocks. Resultant peak forces and resultant average forces (RAF; Figure 2) were significantly greater in the CMJs and 1/2-squats when compared with that in the Blocks (Table 1). The angle of force productions (angle of FP) during the block phases was similar in Blocks, Sled10, and Sled20 (angles 10 ± 3°, 9 ± 2°, and 9 ± 2°, respectively), and there were no statistically significant differences between these values.
Figure 2: Resultant average forces (RAFs; mean and SDs) during the “block phase” in block start (Blocks), sled pulling with a load of 10 and 20% of body mass (Sled10 and Sled20), and during the concentric phase in the countermovement jump (CMJ), CMJ with a load of 10 and 20% of the 1/2-squat 1 repetition maximum (RM) (CMJ10 and CMJ20), 1/2-squat with 70% of 1/2-squat 1RM (S70), and 1/2-squat 1RM (SMAX). *Significantly (p ≤ 0.05) different from Blocks.
Table 1: Force-time variables (mean and SDs) during the “block phase” in block start (Blocks), sled pulling with a load of 10 and 20% of body mass (Sled10 and Sled20), and during the concentric phase in the CMJ, CMJ with a load of 10 and 20% of the 1/2-squat 1 repetition maximum (RM) (CMJ10 and CMJ20), 1/2-squat with 70% of 1/2-squat 1RM (S70), and 1/2-squat 1RM (SMAX).* †
Electromyographic Variables
The activity of the GM was higher in Sled10, Sled20, CMJ10, CMJ20, and 1/2-squats (Table 2; Figure 3A). On the other hand, the activity of the BF was lower, especially in the CMJs and 1/2-squats (Table 2; Figure 3B). The IEMG value of the VL was higher in Sled10 and in 1/2-squats. However, the aEMG values of the VL were quite similar in all exercises (Table 2).
Table 2: Relative IEMG and aEMG values (mean and SDs) during the “block phase” in block start (Blocks), sled pulling with a load of 10 and 20% of body mass (Sled10 and Sled20), and during the concentric phase in the CMJ, CMJ with a load of 10 and 20% of 1/2-squat 1RM (CMJ10 and CMJ20), 1/2-squat with 70% of 1/2-squat 1RM (S70) and 1/2-squat 1RM (SMAX).* †
Figure 3: Relative averaged EMG (aEMG) value of gluteus maximus (A) and biceps femoris (B) (mean and SDs) during the “block phase” in block start (Blocks), sled pulling with a load of 10 and 20% of body mass (Sled10 and Sled20), and during the concentric phase in the countermovement jump (CMJ), CMJ with a load of 10 and 20% of 1/2-squat 1 repetition maximum (RM) (CMJ10 and CMJ20), 1/2-squat with 70% of 1/2-squat 1RM (S70), and 1/2-squat 1RM (SMAX). *Significantly (p ≤ 0.05) different from Blocks.
Kinematic Variables
Table 3 shows peak angular velocities (PAVs) and average angular velocities (AAV), and Table 4 shows minimum (MIN) and maximum angles (MAX) of the hip, knee, and ankle during different exercises. The PAV and AAV of the hip (Figure 4A) and knee (Figure 4B) were mostly significantly lower during the block phase of Sled10 and Sled20. On the other hand, the corresponding values in the CMJ were higher. The PAV and AAV of the knee were also higher in CMJ10 and CMJ20. During S70, all the PAV and AAV values were lower.
Table 3: The PAV and the AAVs (mean and SDs) of the hip, knee, and ankle during the “block phase” in block start (Blocks), sled pulling with a load of 10 and 20% of body mass (Sled10 and Sled20), and during the concentric phase in the CMJ, CMJ with a load of 10 and 20% of 1/2-squat 1RM (CMJ10 and CMJ20) and 1/2-squat with S70.*
Table 4: The MIN and MAX angles of the hip, knee, and ankle (mean and SDs) during the “block phase” in block start (Blocks), sled pulling with a load of 10 and 20% of body mass (Sled10 and Sled20), and during the concentric phase in the CMJ, CMJ with a load of 10 and 20% of the 1/2-squat 1RM (CMJ10 and CMJ20) and the 1/2-squat with 70% of 1/2-squat 1RM (S70).* †
Figure 4: A) Average angular velocity of the hip joint (AAV-HIP) and B) knee joint (AAV-KNEE) (mean and SDs) during the “block phase” in block start (Blocks), sled pulling with a load of 10 and 20% of body mass (Sled10 and Sled20), and during the concentric phase in countermovement jump (CMJ), CMJ with a load of 10 and 20% of 1/2-squat 1 repetition maximum (RM) (CMJ10 and CMJ20) and the 1/2-squat with 70% of 1/2-squat 1RM (S70). *Significantly (p ≤ 0.05) different from Blocks.
Minimum (MIN) and maximum angles (MAX) (Table 4) were mostly similar in the block phase of Blocks, Sled10 and Sled20, although the MIN ankle was smaller in Sled10 and in Sled20. The MIN-hip was larger in CMJ10, CMJ20, and S70 and, on the other hand, MIN-knee and MIN-ankle were smaller in CMJs and in S70. The MAX-hip was greater in CMJ10, CMJ20, and S70, and the MAX-knee was greater in CMJs and in S70.
Associations Between Performance Time in Block Start (10 m) and Force-Time, Electromyographic, and Kinematic Variables of the Studied Exercises
Table 5 shows the statistically significant correlations between performance time in block start (10 m) and force-time, EMG, and kinematic variables in the studied exercises. The highest correlation existed between takeoff velocity in the CMJ and block start (10 m) (r = −0.950, p ≤ 0.001; Figure 5). Takeoff velocity in CMJ10, RPF/body weight (BW) in squats, RAF/BW in CMJs, and relative 1RM in SMAX (load/BM) correlated also significantly with performance time in the block start (10 m). Electromyographic variables did not correlate as significantly as did force-time variables with block start (10 m). The aEMG value of the BF in the CMJ and Blocks, the aEMG value of the VL in Sled10, and the IEMG value of BF in the CMJ correlated significantly with block start (10 m). From the kinematic variables, AAV-hip during CMJ, PAV-hip during CMJ10 and CMJ20, AAV-knee during CMJ and PAV-knee during CMJs correlated significantly with the block start (10 m).
Table 5: Statistically significant correlation coefficients between performance time in block start (10 m) and force-time, EMG, and kinematic variables in measured exercises.*
Figure 5: Relationship between 10-m performance time (seconds) in block start (Blocks) and takeoff velocity (meters per second) in the countermovement jump (CMJ).
Discussion
The primary findings of this study were that the activity of the GM was higher and the activity of the BF was lower in nearly all the studied exercises when compared with that in the block phase of the block start. The additional main findings were that the performance time in the block start (10 m) correlated strongly with CMJs and that the sled pulling and CMJs were generally more velocity- and movement-specific exercises than was the 1/2-squat when compared with the block start. In addition, force production (peak and average force) in the CMJs and 1/2-squats and the angular velocity of the knee in the CMJs were higher than those recorded in the block start. On the other hand, the angular velocities during S70 were significantly lower.
As expected, the performance time in the 10-m run and FPT during the “block phase” was longer in Sled10 and in Sled20. Because of the longer FPTs, also the impulses (IMP) were larger. In turn, the step frequency (mean value of the first 6 steps) was smaller in Sled10 and in Sled20. The results of this study, regarding longer performance time, FPT, and smaller step frequency during the sled pulling, support the findings of the earlier studies (24,26).
During the block phase, the activation of the GM was substantially higher in the sled pulling. The IEMG was about 80% and aEMG was about 40% higher in the sled pulling. The difference in the activation of the BF and the VL was not so obvious, although the IEMG of the VL was significantly higher in Sled10 and aEMG of the BF was significantly lower in Sled20. Although some of the difference in the IEMG of the GM can be explained with a longer FPT; also, aEMG played an essential role. After all, it seems that the GM is substantially more activated and the BF is somewhat less activated in the sled pulling than in the block start. The higher aEMG value could be a result of an increased recruitment, firing rate, and synchronization of motor units (22) in the GM. However, the reason for a higher GM activation and a lower BF activation in the sled pulling as compared with that in the block start remains somewhat unclear, because the movement pattern was very similar in both exercises. Moreover, it is typical that EMG activation increases with strength training and that the increase might be more pronounced at the beginning phases of training (4–8 weeks) (14,15,33). Thus, this increased activation of GM during the sled pulling could in long-term training lead to increased activation during the block start. It can be speculated that increased activation of GM would lead to more efficient force production of GM, which could be beneficial, because the GM is thought to be a very important muscle in sprint running (18).
As expected, nearly all angular velocities were smaller during the block phase of the sled pulling. However, the movement pattern was quite specific to the block start, because the MIN and MAX angles were nearly similar. It can be suggested that Sled10 and Sled20 are quite velocity- and movement-specific when compared with Blocks and, thus, rather a good transfer of training to sport performance could be expected (38). Moreover, it seems that force and EMG adaptations are quite velocity specific. However, there also seems to be transfer from faster to slower velocities, at least, in force production adaptations and, on the other hand, transfer of force production and EMG adaptations from slower to faster velocities seems to be less pronounced or nonexistent (2,7,12,27,32).
Resultant peak force and RAF were larger in the CMJs and both increased as the load increased in CMJs. These results suggest that CMJs overload the force component of the block start. It can be speculated that a long-term usage of the CMJs in training of the sprint start could induce positive changes on the force-velocity curve and consequently improve the performance (7,25). However, when the RPF of the front leg (1,379 N) during Blocks is compared with the RPF in the CMJs divided for each leg (RPF/leg, ∼1,125 N) the force component overloading can be questioned. On the other hand, it is important to notice that the peak force in CMJs is usually produced at the beginning of the concentric phase, and thus, the angles of the leg joints are small. In the block start, peak force of the front leg is produced near the end of the block phase, and consequently, the angles of the leg joints are quite large (leg is quite extended). It can be suggested that the CMJs might overload the force component of the block start, at least, at the beginning of the block phase.
In all the CMJs, the clearest difference in the EMG activity when compared with that in the Blocks was in the activity of the BF. Both the IEMG and aEMG values were substantially lower (42–65%) in the CMJs. On the other hand, in CMJ10 and CMJ20, the aEMG values of the GM were significantly higher. These data suggest that the activation of BF is low and the activation of GM is high in the CMJs as compared with that in the block phase of the block start. The GM is activated in the CMJs in a similar way as in sled pulling and, thus, they could have a similar positive influence on the sprint start if used in the training of sprinters.
Among the kinematic variables, especially the angular velocities of the knee were higher in the CMJs and the mean values of the angular velocities of the hip, knee, and ankle in the CMJs were the highest in the CMJ and the lowest in CMJ20. In the CMJ without a load, also the angular velocity of the hip was higher than that during the block phase of the block start. One reason for the higher angular velocities observed in the CMJs (especially in CMJ without a load) could simply be that force produced toward the ground was produced via both legs, while during the block phase of the block start, approximately the second half of the FPT is used when the force is produced only via the front leg. These results indicate that CMJs are supravelocity exercises for the knee joint when compared with that in the block phase of the block start and CMJ also for the hip joint. On the other hand, according to the PAV-hip and PAV-ankle during CMJ10 and CMJ20, it can be suggested that these exercises were velocity specific with regard to the block phase of the block start. It can be speculated that the usage of CMJs in training of the sprint start could induce positive changes on the force-velocity curve and thus improve the performance (7,25). Moreover, the differences in the MIN and MAX angles were rather small in CMJ, so it can be suggested that the CMJ is also quite a movement specific exercise. Because angular velocities were similar or higher, it can be expected that transfer of force production (and possibly EMG) adaptations to Blocks should be good, if CMJs were used in training (12,32).
Force production time, RPF, and RAF were substantially larger in the 1/2-squats (S70 and SMAX) and the values were larger in SMAX than in S70. The values of the 1/2-squats were also larger than the values of CMJs. It seems that S70 and SMAX significantly overload the force component of the block phase of the block start and, thus, if used in training, could enhance the force component of the force-velocity curve. However, because the angular velocities of the lower limb joints seem to be substantially lower in the squats, it can be speculated that the enhancements on the force-velocity curve occurs especially in lower angular velocities and changes in force production in higher angular velocities might be less pronounced (12,32). Moreover, if the RPF is divided to each leg, RPF/leg in the SMAX was 1,580 N. When this value is compared with the RPF of the front leg during the block phase (1,378 N), the difference is no more that large.
The IEMG values of the GM and VL were remarkably higher in the 1/2-squats. A large amount of the difference can be explained by longer FPTs in the squats, but in the case of SMAX, also the aEMG value of GM was significantly higher. In turn, the aEMG value of BF was remarkably smaller in both S70 and SMAX. It was somewhat surprising that the aEMG value of the GM was not higher in S70 than in CMJ10 and CMJ20, despite the higher force produced in S70. The similarity in the aEMG value of the GM between these exercises could possibly be explained by higher angular velocities (31) of the hip joint and smaller knee (4) and hip angles during CMJ10 and CMJ20. However, as in the sled pulling and CMJs, also in the squats the activation of the GM was higher than in Blocks and, thus, it can be speculated that also the use of the 1/2-squat exercise in training would have a positive influence in the long term on GM activation during the sprint start. In addition, it seems that the aEMG of the GM was higher in SMAX than in the other exercises and, thus, SMAX could be the most efficient of these exercises to induce positive intramuscular neural adaptations (increased motor unit recruitment, firing rate, and synchronization) in long-term training. However, according to earlier studies (2,7,27), transfer of these EMG adaptations to faster contraction velocities seems to be less pronounced.
The angular velocities were 28–68% lower in the S70 than in the block phase of the block start, and although the movement analysis was not done for the SMAX, it is likely that angular velocities were even lower in the SMAX. These results indicate that the 1/2-squat with rather a heavy load is not a very velocity-specific exercise in sprint start training and consequently the transfer of training adaptations to the block start may be compromised (27,32).
Most of the significant correlations existed between CMJs and performance time in the block start (10 m). From the force-time variables, the takeoff velocity in the CMJ correlated most strongly (r = −0.95, p ≤ 0.001). Also, for example, the RPF/BW (S70 and SMAX), RAF/BW (CMJs), and load/BM (SMAX) correlated significantly. According to the results of this study, it seems that the CMJs with rather a light load correlate, at least, somewhat better with the sprint start performance (10 m) than 1/2-squat with a heavy load or sled pulling. However, these findings indicate that high relative force production during CMJs and 1/2-squats is important in achieving a good sprint start. In CMJs, this high relative force production was observed in the high takeoff velocity and, thus, in the high jump height. The takeoff velocity can be calculated from concentric net impulse by dividing the impulse with BM or system mass and, thus, the relative force production (force/BM) is important. In some earlier studies (8,19,37) correlations between the sprint start and speed-strength exercises (different kinds of vertical and horizontal jumps) have been somewhat lower than in this study. However, in the study of Young et al. (39) peak force in relation to body weight (PF/BW) and PF/BW in 100 milliseconds during a concentric jump squat with 19 kg (knee angle 120°) correlated quite strongly with the sprint start (r = −0.86 and r = −0.73, respectively). In the same study, also the maximum force during the isometric squat (knee 120°) correlated quite strongly with the sprint start (r = −0.72, p = 0.07), which was very similar to that shown in this study with RPF/BW of SMAX (r = −0.74). On the other hand, in the study of Harris et al. (16), 1RM squat (knee angle 110°) correlated rather poorly with the sprint start (10 m).
Although the primary findings were, for example, that GM was more activated in nearly all the studied exercises, force production was higher during CMJs and squats, and that the sprint start performance correlated most strongly with the performance in CMJ, we need to be careful when interpreting these results. The reasons for this carefulness include specific problems associated with surface EMG measurements during dynamic movements (moving of muscle in relation to the electrode and changes in the magnitude of the signal with different muscle lengths [13,20]), the cross-sectional study design (interpretations of possible training adaptations have to be considered cautiously), the present subject group (results could be at least somewhat different in different subject populations [e.g., female athletes or elite male sprinters]), and the moderately small n value (because of this especially the correlation values should be interpreted with some caution). However, the influence of the above-mentioned EMG problems might have been small, because the ranges of motions were quite similar in all studied exercises. Moreover, the strength of this study was that the sprint start and different speed-strength and strength exercises were compared extensively (force-time, EMG, and kinematic variables). In earlier studies, the comparison has not been that extensive and, moreover, there has been a lack of studies comparing sprinting and squat type of movements.
In conclusions, nearly in all the studied exercises, the activation of the GM was greater and, on the other hand, the activation of the BF was smaller than during the block phase of the block start. In addition, force production (peak and average force) was higher in the 1/2-squats and CMJs, the angular velocity of the knee was higher in CMJs and the highest correlations existed between CMJs and the performance time in the sprint start. Finally, the transfer of training adaptations might be good from the sled pulling and CMJs to the block start, because of a rather high velocity and movement specificity.
Practical Applications
It should be first emphasized that because of the present cross-section study design, great caution needs to be exercised with regard to definite conclusions for training adaptations. Thus, giving any guidelines or recommendations of the usage of any of the studied exercises in training for the sprint start is somewhat problematic. However, this study gives us information about the relationships between the sprint start and exercises used and based on some previous studies concerning training adaptations, specificity of training adaptations, and EMG activity during the sprint start, some suggestions can be made. Hence, the usage of the sled pulling and CMJs in training can be recommended to induce positive changes in the activation of GM during the block start, because GM activation seems to be significantly higher during the sled pulling and CMJs (CMJ10 and CMJ20). Furthermore, good transfer from the sled pulling and CMJs to the sprint start can be expected because of quite a high specificity in the angular velocities and movement pattern. Countermovement jumps could also induce positive changes at least on the velocity end of the force-velocity curve because of the higher angular velocities, especially at the knee joint, and therefore, enhance the force production at higher velocities and possibly also at angular velocities achieved during the block start. Moreover, the correlation analyses support the usage of CMJs or exercises that improve the CMJ performance in the training of the sprint start, because a good performance in CMJs seems to be strongly associated with a good sprint start performance.
Acknowledgments
The authors would like to thank especially Laura Kajovaara for an essential role in the executing of the measurements. Also, Juha Isolehto and Simon Walker are thanked for their help with the movement analysis and for the familiarization to the EMG measurements, respectively.
References
1. Ahtiainen J, Häkkinen K. Hermo-lihas-järjestelmän toiminnan mittaaminen (Measurement of neuromuscular function). In: Kuntotestauksen Käsikirja (Textbook of Physical Fitness). Keskinen K.L., Häkkinen K., Kallinen M., eds. Helsinki, Finland: Liikuntatieteellinen seura (Finnish Society of Sport Sciences), 2007. pp. 125–193.
2. Andersen LL, Andersen JL, Magnusson SP, Suetta C, Madsen JL, Christensen LR, Aagaard P. Changes in the human muscle force-velocity relationship in response to resistance training and subsequent detraining. J Appl Physiol 99: 87–94, 2005.
3. Blazevich AJ, Jenkins DG. Effect of the movement speed of resistance training exercises on sprint and strength performance in concurrently training elite junior sprinters. J Sports Sci 20: 981–990, 2002.
4. Caterisano A, Moss RF, Pellinger TK, Woodruff K, Lewis VC, Booth W, Khadra T. The effect of back squat depth on the EMG activity of 4 superficial hip and thigh muscles. J Strength Cond Res 16: 428–432, 2002.
5. Cormie P, McBride JM, McCaulley GO. Power-time, force-time, and velocity-time curve analysis during the jump squat: Impact of load. J Appl Biomech 24: 112–120, 2008.
6. Cormie P, McBride JM, McCaulley GO. Power-time, force-time, and velocity-time curve analysis of the
countermovement jump: Impact of training. J Strength Cond Res 23: 177–186, 2009.
7. Cormie P, McGuigan MR, Newton RU. Adaptations in athletic performance after ballistic power versus strength training. Med Sci Sports Exerc 42: 1582–1598, 2010.
8. Cronin JB, Hansen KT. Strength and power predictors of sports speed. J Strength Cond Res 19: 349–357, 2005.
9. Cronin J, McNair PJ, Marshall RN.
Velocity specificity, combination training and sport specific tasks. J Sci Med Sport 4: 168–178, 2001.
10. Delecluse C. Influence of strength training on sprint running performance: Current findings and implications for training. Sports Med 24: 147–156, 1997.
11. Delecluse C, van Coppenolle H, Willems E, van Leemputte M, Diels R, Goris M. Influence of high-resistance and high-velocity training on sprint performance. Med Sci Sports Exerc 27: 1203–1209, 1995.
12. Ewing JL Jr, Wolfe DR, Rogers MA, Amundson ML, Stull GA. Effects of velocity of isokinetic training on strength, power, and quadriceps muscle fibre characteristics. Eur J Appl Physiol Occup Physiol 61: 159–162, 1990.
13. Felici F. Applications in exercise physiology. In: Electromyography: Physiology, Engineering, and Noninvasive Applications. Merletti R., Parker P.A., eds. Hoboken, New Jersey: John Wiley & Sons, 2004. pp. 365–379.
14. Häkkinen K, Alen M, Kallinen M, Newton RU, Kreamer WJ. Neuro-muscular adaptations during prolonged strength training, detraining and re-strength-training in middle-aged and elderly people. Eur J Appl Physiol 83: 51–62, 2000.
15. Häkkinen K, Komi PV. Electromyographic changes during strength training and detraining. Med Sci Sports Exerc 15: 455–460, 1983.
16. Harris NK, Cronin JB, Hopkins WG, Hansen KT. Relationship between sprint times and the strength/power outputs of a machine squat jump. J Strength Cond Res 22: 691–698, 2008.
17. Harris NK, Cronin JB, Hopkins WG, Hansen KT. Squat jump training at maximal power loads vs. heavy loads: Effect on sprint ability. J Strength Cond Res 22: 1742–1749, 2008.
18. Hunter JP, Marshall RN, McNair PJ. Segment-interaction analysis of the stance limb in sprint running. J Biomech 37: 1439–1446, 2004.
19. Holm DJ, Stålbom M, Keogh JWL, Cronin J. Relationship between the kinetics and kinematics of a unilateral horizontal drop jump to sprint performance. J Strength Cond Res 22: 1589–1596, 2008.
20. Kamen G. Electromyographic kinesiology. In: Research Methods in Biomechanics. Robertson D.G.E., Caldwell G.E., Hamill J., Kamen G., Whittlesey S.N., eds. Champaign, IL: Human Kinetics, 2004. pp. 163–181.
21. Kellis E, Arambatzi F, Papadopoulos C. Effects of load on ground reaction force and lower limb kinematics during concentric squats. J Sports Sci 23: 1045–1055, 2005.
22. Komi PV. Training of muscle strength and power: Interaction of neuromotoric, hypertrophic, and mechanical factors. Int J Sports Med 7(Suppl): 10–15, 1986.
23. Kristensen GO, van den Tillaar R, Ettema GJC.
Velocity specificity in early-phase sprint training. J Strength Cond Res 20: 833–837, 2006.
24. Lockie RG, Murphy AJ, Spinks CD. Effects of resisted sled towing on sprint kinematics in field-sport athletes. J Strength Cond Res 17: 760–767, 2003.
25. Malisoux L, Francaux M, Nielsen H, Theisen D. Stretch-shortening cycle exercises: An effective training paradigm to enhance power output of human single muscle fibers. J Appl Physiol 100: 771–779, 2006.
26. Maulder PS, Bradshaw EJ, Keogh JWL. Kinematic alterations due to different loading schemes in early acceleration sprint performance from starting blocks. J Strength Cond Res 22: 1992–2002, 2008.
27. McBride JM, Triplett-McBride T, Davie A, Newton RU. The effect of heavy- vs. light-load jump squats on the development of strength, power, and speed. J Strength Cond Res 16: 75–82, 2002.
28. Mero A. Force-time characteristics and running velocity of male sprinters during the acceleration phase of sprinting. Res Q Exerc Sport 59: 94–98, 1988.
29. Mero A, Komi PV. Reaction time and electromyographic activity during a sprint start. Eur J Appl Physiol Occup Physiol 61: 73–80, 1990.
30. Mero A, Komi PV. EMG, force, and power analysis of sprint-specific strength exercises. J Appl Biomech 10: 1–13, 1994.
31. Moritani T, Stegeman D, Merletti R. Basic physiology and biophysics of EMG signal generation. In: Electromyography: Physiology, Engineering, and Noninvasive Applications. Merletti R., Parker P.A., eds. Hoboken, New Jersey: John Wiley & Sons, 2004. pp. 1–25.
32. Narici MV, Roi GS, Landoni L, Minetti AE, Cerretelli P. Changes in force, cross-sectional area and neural activation during strength training and detraining of the human quadriceps. Eur J Appl Physiol Occup Physiol 59: 310–319, 1989.
33. Reeves ND, Maganaris CN, Narici MV. Plasticity of dynamic muscle performance with strength training in elderly humans. Muscle Nerve 31: 355–364, 2005.
34. Sale DG. Neural adaptation to strength training. In: Strength and Power in Sport. Komi P.V., ed. Oxford, United Kingdom: Blackwell Scientific Publications, 1991. pp. 249–265.
35. Schmidtbleicher D. Training for power events. In: Strength and Power in Sport. Komi P.V., ed. Oxford, United Kingdom: Blackwell Scientific Publications, 1991. pp. 381–395.
36. Slawinski J, Bonnefoy A, Ontanon G, Leveque JM, Miller C, Riquet A, Cheze L, Dumas R. Segment-interaction in sprint start: Analysis of 3D angular velocity and kinetic energy in elite sprinters. J Biomech 43: 1494–1502, 2010.
37. Smirniotou A, Katsikas C, Paradisis G, Argeitaki P, Zacharogiannis E, Tziortzis S. Strength-power predictors of sprinting performance. J Sports Med Phys Fitness 48: 447–454, 2008.
38. Young WB. Transfer of strength and power training to sports performance. Int J Sports Physiol Perform 1: 74–83, 2006.
39. Young W, McLean B, Ardagna J. Relationship between strength qualities and sprinting performance. J Sports Med Phys Fitness 35: 13–19, 1995.
40. Zafeiridis A, Saraslanidis P, Manou V, Ioakimidis P, Dipla K, Kellis S. The effects of resisted sled-pulling sprint training on acceleration and maximum speed performance. J Sports Med Phys Fitness 45: 284–290, 2005.
