In strength training for the upper body, different types of chest press are often used to increase maximal strength. These chest press exercises can be classified by the number degrees of freedom in the exercise. The bench press exercise using a Smith machine is a typical example of a 1 degree of freedom exercise; athletes can only move the barbell straight up and down (vertical movement), whereas bench pressing with a free-weight barbell allows the athlete also to move the barbell horizontally (3 degrees of freedom). When performing a chest press with 2 dumbbells, extra degrees of freedom have to be controlled for because the range of motion at the glenohumeral joint in the frontal and sagittal plane increases together with the possible separate movement in the elbow joint. These differences in degrees of freedom would probably have an influence on the performance (total lifting weight) because more degrees of freedom have to be controlled to be successful.
In maximal and submaximal attempts in bench press, a sticking region (region from the peak velocity until the first local minimal velocity) was found during the upward phase (4,7,12,13), which is often seen as the weakest part in the upward movement at which an attempt potentially will fail (7,10). Elliott et al. (4) suggested that the sticking region is a poor mechanical force position in which the lengths and mechanical advantages of the muscles involved were such that their capacity to exert force was reduced in this region (8).
However, the sticking region has not been studied in other upper-body strength exercises. Strength training exercises with increasing degrees of freedom (e.g., Smith machine, free weights, dumbbells) would probably result in enhanced stress in the neuromuscular system (2). Surprisingly, few studies have compared the effect of increasing degrees of freedom in common resistance training exercises for force output and neural drive (9,14). These studies used weights until 85% of 1 repetition maximum (1RM). Only Sæterbakken et al. (11) compared force output and muscle activity between these 3 exercises with increasing degrees of freedom. The investigators found that differences in 1RM occur with the lowest weight lifted when the most degrees of freedom have to be controlled for (dumbbells). However, no differences in neuromuscular activity in the pectoralis major and anterior deltoid were found, whereas reduced triceps brachii and increased biceps brachii muscle activity in the upward movement during exercises with increasing degrees of freedom were found. Furthermore, during the downward movement, the muscle activity of the biceps, pectoralis, and anterior deltoid was lower when using the Smith machine compared with when using the free weights or dumbbells (11). However, the study only reported different force outputs and the average muscle activation during the whole upward and downward movement of the 3 chest press exercises. When analyzing the downward and upward movements in the different regions, presticking, sticking, and poststicking regions in maximal 1RM attempts, more information about the effect of degrees of freedom can be found.
Therefore, the purpose of this study was to compare the different lifting regions and electromyographic (EMG) activity in maximal 1RM chest press using a Smith machine, a free barbell (conventional bench press), and dumbbells. It was hypothesized that, with increasing degrees of freedom, the sticking region (the weakest region during the lift) would be longer because the muscles need to use a part of their capacity to control this increasing degrees of freedom during the exercise. Furthermore, it was hypothesized that the prime movers (deltoid, triceps, and pectoralis) would have the same muscle activity because it is a maximal performance. However, the biceps muscle activity would increase when the degrees of freedom increases because of the enhanced control of the upward movement.
Experimental Approach to the Problem
This study used a within-subjects crossover design to compare the different lifting regions and EMG activity in 3 chest press exercises with increasing degrees of freedom. The independent variables were the 3 chest press exercises: a bench press in a Smith machine, a free barbell (conventional bench press), and with dumbbells. Dependent variables were the time, velocity, and distance, together with the EMG activity of the pectoralis major, anterior deltoid, triceps brachii, and biceps brachii muscles during downwards, presticking, sticking, and poststicking region in the 3 chest press exercises.
Eleven resistance-trained male sport science students (age 22.6 ± 1.7 years, body mass 78.6 ± 8.0 kg, stature 1.80 ± 0.07 m) with at least 1 year of bench press training experience participated in this study. None of the participants had experience as a competitive power lifter, but all had 4.6 ± 2.2 years of weight-training experience of 2–4 times per week. The participants were familiar with the 3 exercises and performed them regularly as part of their training program. None of the participants had musculoskeletal pain, illnesses, or injuries that might reduce maximal effort during testing. The participants did not perform any additional resistance training exercises that targeted the chest, shoulders, and upper arm muscle groups 72 hours before the test. The study was approved by the local ethics committee, and from all the subjects, a written informed consent was obtained before all testing.
Two weeks before the test, the participants had 3 habituation sessions to identify 1 repetition maximal (1RM) for each of the 3 chest-press exercises (Barbell press, Smith machine, and Dumbbells chest presses). Each familiarization session was separated by 3–5 days. Four to 5 minutes of rest was given between each attempt. Before each habituation session and the experimental test session, a 10-minute warm-up was performed on a cycle ergometer at 75–125 W (60 rpm) followed by 4 warm-up sets: (a) 20 repetitions at 30% of 1RM, (b) 12 repetitions at 50% of 1RM, (c) 6 repetitions at 70% of 1RM, and (d) 1 repetition at 85% of 1RM (11). The percentage of the 1RM was estimated based on the self-reported 1RM of the participants in each of the 3 exercises. A 3-minute rest period was given between each warm-up set (5). At the experimental test, each participant had to conduct a 1RM test in the barbell, dumbbell, and Smith machine chest press. The order of the 1RM test was systematically randomized to avoid the influence of fatigue upon the results.
The position of the hands on the barbell was individually selected, but the forefinger had to be inside the 91-cm mark on the standard Olympic bar that was used. Positioning of the hands was identical in the barbell and Smith machine. In the dumbbells chest press, the position of the arms was individually selected; however, a 2-mm-wide elastic band was placed on each dumbbell to ensure that the dumbbells were lowered to the same height (the elastic band touches the chest) as the other 2 exercises. The hips, shoulders, neck, and head had to be in contact with the bench for each of the exercises, with the feet on the floor, shoulder width apart. Two spotters assisted the participants in the preload phase by lifting the barbell or dumbbells and stabilize the weight until participants had fully extended arms. On audio signal, the participants lowered the barbell to the middle of the sternum until it touched the chest. After lowering the weights to the chest, the participants had to lift the barbell or dumbbells back to the starting position with fully extended elbows. No bouncing of the weights was allowed. Furthermore, the participants had to maintain 5 points (head, thoracic, glutei, feet) of contact on the bench during the lift. If the participant could not lift the dumbbells or barbell to the same vertical position in the upward phase or their technique, the lift was not approved. When the attempt at the assumed 1RM was unsuccessful, the mass was decreased by 2.5 kg per attempt until the real 1RM was found. The participants achieved 1RM for the 3 exercises within 3–5 attempts. A pause of at least 5 minutes between each attempt was used, and the participant had the feeling that he could lift maximally again.
A linear encoder (Ergotest Technology AS, Langesund, Norway) connected to the barbell or dumbbells measured the vertical position and lifting time of the dumbbell or barbell during all 3 exercises with a 0.075-mm resolution and counted the pulses with 10 millisecond intervals (1). The vertical displacement was measured in relation to the lowest point of the barbell (zero distance). Velocity of the barbell was calculated by using a 5-point differential filter with software Musclelab V8.10 (Ergotest Technology AS). The linear encoder was synchronized with the EMG recordings using a Musclelab 3010e and analyzed by software V8.10 (Ergotest Technology AS).
Before the 1RM experimental test, the skin was prepared (shaved, washed with alcohol, abraded) for placement of gel coated surface EMG electrodes. Electrodes (11-mm contact diameter) were placed on the dominant side of the body on the belly of the muscle in the presumed direction of the underlying muscle fibers with a center-to-center distance of 2.0 cm according to the recommendations by SENIAM (6). Self-adhesive electrodes (Dri-Stick Silver circular sEMG Electrodes AE-131, NeuroDyne Medical, Cambridge, MA, USA) were positioned on the belly of the pectoralis major (sternocostal head), the anterior deltoid, the triceps brachii (lateral head), and biceps brachii (short head) (11). To minimize noise induced from external sources, the raw EMG signal was amplified and filtered using a preamplifier located as near to the pickup point as possible. The EMG signals were sampled at a rate of 1,000 Hz. The signals were band pass filtered with a cut-off frequency of 8 and 600 Hz, after which the root mean square (RMS) was calculated. The RMS-converted signal was resampled at a rate of 100 Hz using a 16-bit A/D converter with a common mode rejection rate of 106 dB. The stored data were analyzed using commercial software (Musclelab V8.10, Ergotest Technology AS).
To locate possible differences in muscle activity during the 1RM bench press movement, the average RMS was calculated for each of the 4 regions. The first region was from the highest downward velocity point (vdownwards) to the lowest barbell point where the velocity is zero (v0): downward region. The second region is from the lowest barbell point until the from the lowest barbell point until the maximal barbell velocity (vfirst peak): the presticking region. The third region was from the maximal barbell velocity until the first located lowest barbell velocity (vmin): the sticking region. The last period, the poststicking region, started at vmin to the second maximal barbell peak velocity (vsecond peak), which was also called the strength region (Figure 1) (7).
To assess the differences in 1RM, different regions and neuromuscular activity in these 4 regions during the 3 chest-press exercises, a repeated 3 (exercise: smith machine, barbell, dumbells) × 4 (region: downwards, presticking, sticking, and poststicking) analysis of variance (ANOVA) design was used on the EMG data of the 4 muscles. On the kinematics (load, time, velocity, and distance) a 1 × 3 (chest press exercise) repeated-measures ANOVA was used. The least significant difference post hoc analyses were conducted to determine pairwise differences. All the results are presented as mean ± SD. In the case that the sphericity assumption was violated, the Greenhouse-Geisser adjustments of the p values are reported. The level for significance was set at p < 0.05. Statistical analysis was performed in SPSS version 18.0 (SPSS, Inc., Chicago, IL, USA).
The maximal lifted weight at 1RM among the 3 chest-press exercises was significantly different. The participants achieved the highest 1RM strength using the free barbell (106.4 ± 15.5 kg), followed by the Smith machine (103.6 ± 14.8 kg) and dumbbells (89.5 ± 13.7 kg). The intraclass correlation coefficients (calculated from the familiarization test and the experiment test) for the 1RM exercises were 0.95 (Smith), 0.95 (barbell), and 0.86 (dumbbells).
All 3 exercises showed that all the 4 regions and the total lifting time (Smith machine: 2.54 ± 0.59 seconds, barbell: 2.21 ± 0.48 seconds, dumbbells: 2.56 ± 0.98 seconds) were the same between the 3 chest press exercises (p = 0.33). However, the time length of the different regions was significantly different between the exercises (p = 0.032). The presticking region was significantly shorter for the barbell exercise and the Smith machine compared with that for the dumbbells (Figure 2A). The time in the sticking region was significantly longer when lifting the barbell compared with when lifting the dumbbells, whereas in the poststicking region, the time was the significantly shorter for the barbell exercise compared with the lift in the Smith machine (Figure 2A).
The maximal velocity in the presticking region was significantly higher when using dumbbells compared with when doing the other 2 chess press exercises (p ≤ 0.023; Figure 2 B). This resulted in a higher vertical height between the weights and the sternum at this maximal velocity point and the heights at the minimal velocity and the second peak velocity for the dumbbells (Figure 2C). However, when just the intervals between the different heights at these different points were taken, only a significant difference (p ≤ 0.002) at the first maximal velocity point was found for the dumbbells with the other 2 exercises.
Muscle Activity between the Three Different Exercises
The muscle activity of the pectoralis muscle during the downwards and the presticking region of the chest press in the Smith machine was significantly lower than in the other exercises (p ≤ 0.043; Figure 3), whereas the deltoid muscle activity only was significantly lower between the chess press in the Smith machine compared with the barbell in the sticking and the poststicking region (p ≤ 0.028; Figure 3). The muscle activity of the biceps was significantly lower in the downwards and the presticking region when pressing in the Smith machine compared with the other 2 exercises (p ≤ 0.042), whereas in the sticking and the poststicking regions, the chess press with the dumbbells showed significantly higher muscle activation compared with the other 2 exercises (p ≤ 0.019; Figure 4). The triceps activity was significantly lower in all regions with the dumbbells chess press exercise when compared with the other 2 exercises (p ≤ 0.048), and no significant differences in the muscle activity between the barbell and the Smith machine were found (p ≥ 0.13; Figure 4).
Muscle Activity Development over the Different Regions
For the pectoralis muscles, only a significant decrease was found in the dumbbells exercise from the downwards and the presticking regions to the sticking and poststicking regions (p ≤ 0.036), whereas in the other exercises, no significant changes in the pectoralis muscles were observed (p ≥ 0.085; Figure 3). The deltoid muscles showed increased muscle activation in all 3 exercises from the downward region to the lifting regions (p ≤ 0.014). Furthermore, significant increased muscle activity was found for the dumbbells and the barbell chess presses from the presticking to the sticking region (p ≤ 0.038; Figure 3).
For the biceps muscle, the activity decreases from the presticking region to the sticking region from both the barbell (p = 0.002) and the Smith machine chess press (p = 0.003), whereas no changes occur in muscle activity with the dumbbells (p = 0.113; Figure 4). The triceps activity in all 3 exercises increased from the downwards to the presticking region (p ≤ 0.029) and from the presticking region to the sticking (p ≤ 0.002) and the poststicking region (p ≤ 0.017), with no significant differences in muscle activity between the sticking and poststicking region (p ≥ 0.09; Figure 4).
The main aim of this study was to investigate the effect of increasing degrees of freedom in maximal chest pressing exercises upon the different lifting regions and EMG activity. Differences were found in the different lifting regions between the exercises together with differences in muscle activity. However, the differences found did not follow the line of increasing degrees of freedom that would result in a longer sticking region.
The 1RM was the highest when lifting with the barbell, followed by the smith machine and the dumbbells. This was in line with the findings in earlier studies of Cotterman et al. (3) and Sæterbakken et al. (11) who also found that subjects could lift more weight with the conventional barbell compared with the lifts in the Smith machine. An explanation for this difference could be that in this study, the participants in their daily training preferred free barbell, and the results could be a training-specific adaptation. It was hypothesized that by increasing the degrees of freedom, the sticking region would be longer because the muscles have to control for more degrees of freedom. However, the sticking region was larger when lifting the conventional barbell compared with when lifting the dumbbells (Figure 2A). Furthermore, the poststicking region was shorter with the barbell compared with the lift in the Smith machine, whereas the time of the presticking region was the longest with the dumbbells compared with the other 2 exercises (Figure 2A). These findings indicate that increasing degrees of freedom did not follow the hypothesis.
While lifting the dumbbells, the participants had a longer time in the presticking region, they could produce a longer acceleration and thereby a higher first peak velocity (Figure 2B). This longer acceleration and higher peak velocity resulted in a higher vertical position of the dumbbells compared with the other 2 exercises (Figure 2C), which also resulted in higher vertical heights at minimal velocity and second peak velocity points (Figure 2C).
There was no clear finding in accordance to the hypotheses that with increasing degrees of freedom, the muscle activity of the antagonist muscles would increase and that the prime movers would be the same because the performance would be maximal. Only the muscle activity in the biceps and pectoralis was lower in the downward region when lifting in the Smith machine compared with the other 2 exercises (Figures 3 and 4). The muscle activity in the sticking and poststicking region was higher for the biceps in the dumbbells lift compared with the others (Figure 4). Furthermore, the deltoid muscle activity was lower for lifts in the Smith machine with the barbell in these 2 regions (Figure 3), whereas the triceps activity was lower in all the regions when lifting with dumbbells compared with that in the other 2 exercises (Figure 4).
The lower pectoralis activity in the downwards and presticking region in the Smith machine (Figure 3) can be explained by the path of the barbell: In the Smith machine, the barbell is moved straight up and downwards, whereas in lifting free weights, the barbell is probably moved more to the belly as found in earlier studies upon free-weight bench press (4,8,13). Thus, the barbell is also moved horizontally when lowered. The horizontal displacement from the starting point to the lowest point of the barbell shown by van den Tillaar and Ettema (12) and Madsen and McLaughlin (8) is around 0.12 m, which would probably cause higher activity of the pectoralis muscles compared with the straight up and downward movement. During the sticking and poststicking regions, these horizontal differences will be less because the upward movement is trying to move back to the same horizontal position as it started the exercise. Furthermore, it was found that the deltoid muscle activity was significantly lower between the chest press in the Smith machine compared with the barbell in the sticking and the poststicking regions (Figure 3), whereas no differences were found between the barbell and the dumbbells for this muscle. It could be that during the sticking and poststicking region the deltoid muscle takes over a part of the function of the pectoralis (abduction of the arm) that occurs in the dumbbells and the barbell exercise because of the extra degrees of freedom. An indication of taking over the function is the increased activity of the deltoid muscle from that presticking to the sticking region in the dumbbells and free-weight exercise (Figure 3). However, 3D kinematic analysis has to be performed, which can underline these statements.
The fact that no differences were found between barbell and dumbbells was because the deltoid muscle is not active in extension of the elbow that can occur with dumbbells press as extra degrees of freedom. The muscle activity of the deltoid in this study was a bit different from the activities found by Sæterbakken et al. (11). They found a difference in the downward phase but not in the concentric phase, whereas in this study, we found no differences in the downward region and significant differences in the sticking and poststicking regions between the 3 exercises. A simple explanation is that we divided the 2 phases into 4 different regions, whereas in the downward region, only the muscle activity was averaged from the highest downward movement to the lowest barbell point in our study when compared with Sæterbakken et al. (11) who averaged over the whole downward movement. In the upward movement, we averaged over 3 regions, whereas Sæterbakken et al. (11) took the whole movement including the deceleration region (from vsecond peak to the end of the lift), which could influence the total muscle activity of the deltoid.
The muscle activity of the biceps was significantly lower in the downwards and the presticking region when pressing in the Smith machine compared with the other 2 exercises, whereas in the sticking and the poststicking regions, the chess press with the dumbbells showed significantly higher muscle activation compared with that in the other 2 exercises (Figure 4). The short head of the biceps attaches to (originates from) the coracoid process at the top of the scapula, which results in an abduction or adduction function that occurs during the dumbbells and free-weight exercise. During lowering in the Smith machine, the biceps does not have to control abduction movement in the shoulder that occurs with the dumbbells and the conventional bench press.
In the upward movement in the dumbbells exercise, the extra degrees of freedom rotation in the shoulder joint and flexion and extension of the elbow have to be controlled. To control the flexion-extension movement, the biceps has to activate more, prohibiting that the elbow does not extent too much or too quick. Otherwise, this extension could cause a higher moment arm at the shoulders. Therefore, higher activity in the biceps during the sticking and poststicking region in the dumbbells exercise was observed compared with the other 2 exercises.
This extra degree of freedom (flexion and extension of the elbow) also influenced the triceps activity during the four regions in the dumbbells exercise. The triceps activity was significantly lower with the dumbbells chest press exercise when compared with the other 2 exercises (Figure 4). In the Smith machine and the conventional barbell lift, this degree of freedom does not have to be controlled. Thus, when the triceps would be more active, the dumbbells would move more laterally from the body thereby creating a larger moment arm on the elbow and shoulder joint, which would result in more activity of the other muscles and a lower performance. By decreasing the triceps activity and increasing the biceps activity (in the sticking and poststicking region), this outward movement would be less.
However, it should be noted that we used only 1 electrode pair on each of the muscles (pectoralis, deltoid, biceps, and triceps). Because these muscles have multiple heads, other parts of these muscles could be active during the different lifts and thereby influence the differences observed in our study. Furthermore, surface EMG has inherent technical limitations and can provide only an estimate of neuromuscular activation. There is an inherent risk of crosstalk from neighboring muscles. Even if a small interelectrode distance was used, muscles could have been affected by surrounding muscles during EMG recordings. Further research should investigate the neuromuscular pattern in increasing degrees of freedom when the total weight is the same in each exercise. This makes it possible to compare how much activity each muscle have to deliver when only the degrees of freedom changes and not also the weight.
About the muscle activity during the different exercises, it was observed that for the pectoralis muscles, only a significant decrease was found in the dumbbells exercise from the downwards and the presticking regions to the sticking and poststicking regions, whereas in the other exercise, no significant changes in the pectoralis muscles were observed (Figure 3). In earlier studies of van den Tillaar and Ettema (12,13), no differences in muscle activity for the pectoralis muscle between the different regions during maximal barbell chest press were found. The differences in pectoralis activity during the 4 regions in the dumbbells chest press is probably caused by the increased activity during the downwards and presticking regions even if there were no significant differences found with the barbell chest press (Figure 3).
The deltoid muscles showed increased muscle activation in all 3 exercises from the downward region to the lifting regions, which was in line with the earlier studies of van den Tillaar and Ettema (12) and Sæterbakken et al. (11) who also found significant lower activity in the downwards compared with that in the upward regions. This indicates that the deltoid muscle is not so active in slowing down the weight downwards but more as a prime mover when lifting the weights.
The biceps muscle activity decreases from the presticking region to the sticking region in both the barbell and the Smith machine chess press, whereas no changes occur in muscle activity with the dumbbells (Figure 4). This was in line with the earlier findings of van den Tillaar et al. (12,13) and Sæterbakken (11) who showed that the highest biceps activity during barbell exercises occurs during the downward region and that the main function of the biceps is as stabilizer. The fact that no differences in activity were found for the biceps muscles during the dumbbells exercise in the different regions of the upward movement is probably because of controlling the extra degree of freedom during the lift (flexion and extension of the elbow).
The triceps activity in all the 3 exercises increased from the downwards to the presticking region and from the presticking region to the sticking and the poststicking region, with no significant differences in muscle activity between the sticking and poststicking region (Figure 4). This was also found in earlier studies of van den Tillaar et al (12), who observed the lowest triceps activities during the downward region. It shows clearly that the triceps in all 3 exercises behaves the same and that the triceps in all the exercises is one of the prime movers to lift the weight.
Differences were found in the different lifting regions between the exercises together with differences in muscle activity. However, the differences found did not follow the line of increasing degrees of freedom that would result in a longer sticking region. Thus, training in chest press using a Smith machine, a free barbell (conventional bench press), and dumbbells exhibits different muscle use during the exercises. Therefore, it is possible to choose to train a particular chest press exercise with the purpose of training a particular muscle more than the others (e.g., the triceps is more active using a Smith machine or a free barbell than with the dumbbells exercise). The different chest press exercises can increase and decrease the neuromuscular activity of the involved muscles. The information gained from our study can help trainers and athletes in their understanding about the limitations of the muscles as a result of the effect of degrees of freedom. We recommend that dumbbell chest presses, with the greatest degrees of freedom, can be executed as a supplement to ordinary bench press.
1. Arnason A, Sigurdsson SB, Gudmundsson A, Holme I, Engebretsen L, Bahr R. Risk factors for injuries in football. Am J Sports Med 32: 5–16, 2004.
2. Behm DG, Anderson KG. The role of instability with resistance training. J Strength Cond Res 20: 716–722, 2006.
3. Cotterman ML, Darby LA, Skelly WA. Comparison of muscle force production using the Smith machine and free weights for bench press and squat exercises. J Strength Cond Res 19: 169–176, 2005.
4. Elliott BC, Wilson GJ, Kerr GK. A biomechanical analysis of the sticking region in the bench press. Med Sci Sports Exerc 21: 450–462, 1989.
5. Goodman CA, Pearce AJ, Nicholes CJ, Gatt BM, Fairweather IH. No difference in 1RM strength and muscle activation during the barbell chest press on a stable and unstable surface. J Strength Cond Res 22: 88–94, 2008.
6. Hermens HJ, Freriks B, Disselhorst-Klug C, Rau G. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol 10: 361–374, 2000.
7. Lander JE, Bates BT, Swahill JA, Hamill J. A comparison between free-weight and isokinetic bench pressing. Med Sci Sports Exerc 17:344–353, 1985.
8. Madsen N, McLaughlin T. Kinematic factors influencing performance and injury risk in the bench press exercise. Med Sci Sports Exerc 16: 376–381, 1984.
9. McCaw ST, Friday J. A comparison of muscle activity
between a free weight and machine bench press. J Strength Cond Res 8: 259–264, 1994.
10. Newton R. Biomechanics of conditioning exercises. In: Conditioning for Strength and Human Performance. Chandler T. J., Brown L. E., eds. Baltimore, MD: Wolters Kluwer Health, 2008. pp. 77–93.
11. Sæterbakken A, van den Tillaar R, Fimland M.A. comparison of muscle activity
and 1-RM strength of three chest-press exercises with different stability requirements. J Sports Sci 29: 1–6, 2011.
12. van den Tillaar R, Ettema G. A comparison of kinematics and muscle activity
between successful and unsuccessful attempts in bench press. Med Sci Sports Exerc 41: 2056–2063,2009.
13. van den Tillaar R, Ettema G. The “sticking point” in bench press. J Sports Sci 28: 529–535, 2010.
14. Welsch EA, Bird M, Mayhew JL. Electromyographic activity of the pectoralis major and anterior deltoid muscles during three upper-body lifts. J Strength Cond Res 19: 449–452, 2005.