One of the controversies when prescribing resistance training programs is whether the use of free weights or machines is better for building muscle mass and strength. Free weights provide isotonic resistance, which provides the same amount of resistance throughout the range of motion. This does not always match the strength curve of the muscle; that is, the constant resistance throughout the range of motion offered by free weights does not always match the strength of a muscle, which varies throughout the range of motion. Some machines use “cam” pulley systems, elastics, hydraulics, and pneumatic resistance, which may better match the strength curves of typical movements.
Despite these potential benefits of training with machines, training with free weights may allow for greater recruitment of muscle mass. A major difference between training with free weights and machines is that training with most machines provides a very stable environment, whereas training with free weights requires more stabilization and balance, which may result in greater recruitment of muscle. Using free weights compared with more stable machines results in greater muscle activation (as measured by electromyography) during upper- (15) and lower-body (25) strength exercises. The increased muscle recruitment during free-weight activities can potentially provide a more anabolic stimulus. For example, acute training sessions with free-weight squat exercise results in greater release of anabolic hormones such as free testosterone and growth hormone compared with the more stable leg press exercise (26). This greater anabolic hormone response could potentially lead to greater muscle hypertrophy and strength over time (21).
The purpose of our study was to compare the effects of training with only free weights or machines on anabolic hormone response, measured by free testosterone levels, muscle mass, and strength. This was performed by randomly dividing men and women into 2 different training groups: one that trained exclusively with free weights and the other that trained exclusively with machines for 8 weeks. The hypothesis was that free-weight training would result in greater increases in free testosterone concentration, muscle mass, and strength compared with training with machines over 8 weeks.
Experimental Approach to the Problem
Subjects were randomly assigned to either a free-weight or machine training group after stratifying by sex, months of training experience, and whether they used mostly free weights, mostly machines, or an equal mix of both during their typical training. The total duration of the exercise study was 8 weeks. Hormone levels were assessed using saliva samples collected before and after workouts at the beginning, midway (4 weeks), and end of the study (8 weeks). Body composition, muscle thickness, and strength were measured during the week before the training intervention and during the week after the training intervention.
Ethical approval was obtained from the University of Saskatchewan biomedical review board for research in humans, and subjects were informed of the benefits and risks of the investigation before signing an institutionally approved informed consent document to participate in the study. Forty-six healthy subjects volunteered for this study (20 men and 26 women, age range 18–30 years) and were randomized to machine or free-weight training groups using a computerized random number generator. The free-weight group did not differ compared with the machine group for baseline age (23 ± 4 vs. 22 ± 3 years), mass (67 ± 8 vs. 74 ± 6 kg), height (172 ± 10 vs. 171 ± 10 cm), or previous resistance training experience (27 ± 25 vs. 26 ± 24 months).
Lean Tissue Mass
Lean tissue mass was measured before and after the exercise program by air displacement plethysmography (BOD POD: Life Measurement Instruments, Concord, CA) using methods previously described in detail (4). Our laboratory has a coefficient of variation of 0.80% for lean tissue mass using the BOD POD and a correlation coefficient of 0.98 when comparing BOD POD with dual-energy X-ray absorptiometry.
Muscle thickness was measured at the quadriceps and biceps before and after the exercise program with B-mode ultrasound (Aloka SSD-500, Tokyo, Japan), as previously described (4). The coefficients of variation for these measurements in our laboratory are 0.9 and 2.6% for the quadriceps and biceps, respectively.
Strength was assessed by performing a 1-repetition maximum (1RM) on a free-weight bench press, 6–10RM free-weight squat, 1RM Smith machine bench press, and a 6–10RM Smith machine squat. The free-weight strength tests were performed at least 2 days apart from the Smith machine strength tests. The order in which subjects performed their bench presses and squats and the mode in which they were tested on first were randomized. A predicted 1RM was determined based on the 6–10RM value for the squat exercises (12) for safety reasons. The Life Fitness Smith machine consisted of an Olympic bar that has each end attached to an upright rail. The free-weight bench press was performed using a barbell and flat bench press. The free-weight squat used a power rack and a barbell. For all exercises, subjects warmed up using a light weight of their choice, and they then performed up to 5 trials for a maximal lift. There was 3–5 minutes of rest given between each trial.
During the free-weight bench press 1RM, hands were placed approximately shoulder-width apart, feet on the floor, and back against the bench. The subject received help unracking the bar, and they lowered the bar until it contacted their chest at which point they pushed the bar back up to full elbow extension where they received help reracking the bar. If the subject was unsuccessful, a spotter helped rerack the bar. For the free-weight squat 6–10RM, subjects' feet were approximately shoulder-width apart. The subject received help unracking the bar, and they squatted down until their knees were approximately at 90° where they stood back up until full hip extension was achieved. Once they had reached their 6–10RM, they received help reracking the weight. Depth of each repetition was controlled for by attaching a TheraBand between the frames at a height that when the bar touched the band at the bottom range of motion, the subject was at a 90° knee angle. Once the bar touched the band, the subject received a verbal cue to stand back up. The height of the TheraBand was recorded for the post-test strength assessment. If the subject could not complete the repetition, they lowered the bar onto the safety rails.
For the Smith machine bench press 1RM, the subject received help unracking the bar by slightly rotating the safety hooks off of the latches located on the frame of the machine and lowered the bar until it contacted their chest and then pushed the bar back up to full elbow extension where they received help reracking the bar by slightly rotating the safety hooks back onto the latches. If the subject was unsuccessful, a spotter helped rerack the bar. For the Smith machine squat 6–10RM, subjects' feet were approximately shoulder-width apart. The subjects received help unracking the bar, and they squatted down until their knees were approximately at 90° where they stood back up until full hip extension was completed. Once they had completed their 6–10RM, they received help reracking the weight (same as the Smith machine bench press). Depth of each repetition was controlled for by placing a box on the outside of the frame and stacking mats high enough that when the bar touched the mat at the bottom range of motion, the subject was at an approximately 90° knee flexion angle. Once the bar touched the mat, the subject received a verbal cue to stand back up. The height of the box and mats was recorded for post-test strength assessments. If the subject could not complete the repetition, they lowered the bar onto the safeties. The coefficients of variation for the free-weight and Smith machine bench press and squat exercises in our laboratory ranged from 5 to 8% (intraclass correlation coefficients = 0.90–0.95).
A standardized workout was performed at the first, midpoint (4 weeks), and last workout (8 weeks). This workout consisted of performing only the bench press and squat on their designated mode of training. These 2 exercises consisted of performing 4 sets of 6–10 repetitions with 1.5 minutes of rest between sets. Loads for the first hormone collection workout were calculated as 70% of 1RM based on their pre-test strength assessments. Loads for the midway and final workout were based on the weights being used during the workouts just before the hormone assessment day. Salivary samples were collected before the start of these 3 workouts and 15 minutes after the workouts. Salivary hormone levels reflect the free plasma concentration and bioactive component of steroid hormones (10). Time of the day was recorded for the first workout so that the midway workout and the final workout were performed at the same time of the day. This is important because of the circadian rhythm that affects free testosterone and free cortisol levels (9,10). Subjects were asked to ensure they were fully hydrated before hormone assessment sessions. A food record was recorded one day before the hormone collection so that the same food could be ingested on each of the days before the next 2 hormone collection days. To minimize the effect of recent exercise, subjects were told not to exercise for 2 hours before their hormone collection sessions. Salivary free testosterone and free cortisol were measured using enzyme-linked immunoassay kits according to the manufacturer's instructions (Salimetrics, State College, PA). Saliva was collected from passive drool through a short straw and into a polypropylene vial. Samples were frozen at −30° C until analysis. Once thawed, the saliva samples were pipetted into the appropriate wells, mixed on a plate rotator for 5 minutes at 500 rpm, and incubated in the dark at room temperature for an additional 25 minutes. The samples were read in a plate reader at 450 nm. Three samples were taken, and an average value was calculated. Our laboratory had intra-assay coefficients of variation ranging from 4.0 to 7.2% for free cortisol and 4.6–8.6% for free testosterone. All samples for a given individual were assessed within the same assay so that individual results were not affected by inter-assay variability.
The exercise program lasted for 8 weeks and consisted of a 2 days on and one day off cycle (e.g., half the muscle groups were trained on one day and half the next day, followed by a day of rest). Eight weeks was chosen as it is an adequate amount of time to realize hypertrophy and increases in strength for most muscle groups (30). Day one trained the chest, back, and triceps. The free-weight exercises included the flat barbell bench press, incline barbell bench press, bent over barbell row, chin-ups, supine elbow extension, and dumbbell kickbacks. The machine exercises were performed on Technogym (Seattle, WA), Hammer Strength (Cincinnati, OH), Life Fitness (Schiller Park, IL), and APEX (Saanichton, BC, Canada) equipment. The Technogym equipment uses a cam pulley system, which is designed to match the strength curve of the specific movement. The machine exercises for the chest, back, and triceps included the Smith machine (Life Fitness) bench press, Smith machine incline bench press, Hammer Strength–seated row, Technogym lat pulldown, Technogym machine triceps press-down, and rope press-down (Life Fitness pulley system). Day 2 trained the legs, shoulders, and biceps. Free-weight exercises included the squat, straight leg dead-lift, lunge, single-leg calf raise, dumbbell shoulder press, dumbbell lateral raise, camber bar curl, and preacher curl. The machine exercises for the legs, shoulders, and biceps included the Smith machine squat, Technogym quadriceps extension, Technogym-seated hamstring curl, APEX machine calf raise, Technogym machine shoulder press, Technogym machine lateral raise, Technogym machine biceps curl, and Hammer Strength machine preacher curl. For the first 3 weeks, all exercises were performed for 4 sets of 8–10 repetitions with 1 minute of rest between sets. For the next 3 weeks, weight was increased, and all exercises were performed for 4 sets of 6–8 repetitions with 1.5 minutes of rest between sets. For the last 2 weeks, weight was increased again, and all exercises were performed for 3 sets of 4–5 repetitions with 2 minutes of rest between sets. Intensity was increased throughout the program once a subject was able to complete their required repetitions with good form, to achieve progressive overload. The exercise program also increased load as volume decreased to mimic a taper effect, which has been shown to promote strength increases (7). If the subject performed a set outside of the desired repetition range, they were instructed to adjust the weight for the following sets so that they would complete the appropriate number of repetitions required. This allowed for an adequate progression in training load. Workouts were recorded in a detailed activity log. All workouts took place in our university's fitness center where fully qualified exercise professionals (i.e., Canadian Society for Exercise Physiology Certified Exercise Physiologists or Certified Personal Trainers) were available to provide assistance during the workouts. Over the 8 weeks, there were approximately 38 training sessions; therefore, with the split-body program, each muscle group was trained over 19 sessions.
A 2 × 2 × 2 mixed (between–within) analysis of variance (ANOVA) was conducted with group (free-weight group vs. machine group), sex (men vs. women), and time (baseline vs. 8 weeks) as factors to determine the differences between groups for lean tissue mass, muscle thicknesses, and strength over time.
A 2 × 2 × 2 × 3 mixed (between–within) ANOVA was conducted with group (free-weight group vs. machine group), sex (male vs. female), time during workout (pre vs. post), and time of the training program (baseline vs. 4 vs. 8 weeks) as factors to determine the difference between groups for hormone levels during workouts and over time. Tukey's post hoc tests were run when significant interactions were found. All statistical analyses were conducted using STATISTICA 7.0 (Tulsa, OK). All values are expressed as means ± SD, except in graphs, where standard errors were used for clarity. A p value of 0.05 was accepted as significant.
The subject flow through the study is indicated in Figure 1. Fifteen men and 21 women completed the study (7 men and 11 women in the free-weight group, 8 men and 10 women in the machine group). The reason for dropout was the time commitment needed to complete the workouts. There was no difference between groups for compliance to the training with the free-weight group attending 83 ± 29% and the machine group attending 81 ± 12% of their training sessions. As an indication of the subjects' training experience, the ratio of their squat and bench press strength to body mass at baseline was as follows: machine-group men: Smith machine bench press 1.0 ± 0.2, Smith machine squat 1.6 ± 0.5, free-weight bench press 1.0 ± 0.2, and free-weight squat 1.6 ± 0.4; machine-group women: Smith machine bench press 0.7 ± 0.1, Smith machine squat 1.7 ± 0.2, free-weight bench press 0.7 ± 0.1, and free-weight squat 1.6 ± 0.3; free-weight group men: Smith machine bench press 1.2 ± 0.3, Smith machine squat 2.1 ± 0.2, free-weight bench press 1.1 ± 0.2, and free-weight squat 2.0 ± 0.2; free-weight group women: Smith machine bench press 0.6 ± 0.1, Smith machine squat 1.5 ± 0.5, free-weight bench press 0.6 ± 0.1, and free-weight squat 1.5 ± 0.4.
Lean Tissue Mass and Muscle Thickness
There were no differences in lean tissue mass over time or between groups (Table 1 and Figure 2A). Biceps and quadriceps muscle thickness increased over the training program (time main effect p < 0.01), with no differences between groups (Table 2 and Figure 2B, C). There was a sex main effect (p < 0.01), for lean tissue mass, and biceps and quadriceps muscle thickness with higher values in men compared with women, as would be expected. There were no sex × time interactions, indicating men and women responded similarly to the training program.
There was a group × time interaction (p = 0.05) for the machine bench press with the machine training group experiencing a greater increase in machine bench press strength compared with the free-weight training group (Table 3 and Figure 3A). There were no other differences between groups over time for any other strength measure (Figure 3B–D). There were significant time main effects for free-weight bench press, free-weight squat, and machine squat (p < 0.01), with strength increasing from before to after training. There was a significant sex main effect (p < 0.01) for all strength tests, with men higher than women, as would be expected. There were no sex × time interactions for any strength measures, indicating men and women responded similarly to the training program.
All hormone results are presented in Table 4. There was a significant group × sex × time during workout interaction for free testosterone (p < 0.05). Tukey's post hoc analyses indicated that only the free-weight training men significantly increased free testosterone from before to after workouts (p < 0.01; Figure 4). There was no significant change in free cortisol at any time point for either sex or training group. There was a significant sex × time during workout interaction for the free testosterone to free cortisol ratio (p < 0.05). Only men had significant increases in the free testosterone to free cortisol ratio during workouts, increasing from 7.0 ± 3.7 to 8.8 ± 5.3 (p < 0.01). There were no changes over the duration of the 8 weeks of training in any hormone measure (i.e., there were no “chronic” changes in any of the hormone measures).
The major finding of this study is that free weight and machine training were equally effective for increasing muscle thickness and strength. These findings do not support our hypothesis that training with free weights would result in greater gains in muscle mass and strength. The second major finding is that the men training with free weights experienced a significant acute increase in free testosterone from before to after workouts. This finding partially supports our hypothesis that the group training with free weights would have greater increases in anabolic hormone response. The hypotheses were based on the evidence that training with free weights activates more muscle mass (1,15,25), which should cause a greater increase in free testosterone (26), and over time causes a greater increase in muscle mass and strength (21).
We found no significant changes for either training group in whole-body lean tissue mass assessed by the Bod Pod but significant increases in biceps and quadriceps muscle thickness across both groups. Subjects had previous training experience and might have been close to their ceiling level of lean body mass, and 8 weeks of resistance training may not have been enough to induce a further increase. Our results are in agreement with one other study that previously used the Bod Pod to assess changes in lean tissue mass. Rossi et al. (22) compared training with free-weight squat vs. machine-based leg press for 10 weeks and found no differences between groups for changes in lean tissue mass. Similarly, Maddalozzo and Snow (14) found that 24 weeks of training with a seated resistance training program or standing free-weight program produced equivalent increases in lean body mass as measured using dual-energy x-ray absorptiometry. Their free-weight program however also included some machine-based exercises; therefore, their programs were not exclusively free weight or machine-based. Boyer (2) compared 3 different training modes consisting of 2 different types of machines or free weights and found no significant differences for changes in body composition as assessed by skinfolds and girths between groups over time. Overall, these results support the contention that free weight and machine-based training are equally effective for increasing muscle mass.
Both the free-weight training group and the machine training group had significant increases in free weight and Smith machine squat strength and free weight and Smith machine bench press strength. These findings do not support our hypothesis that the free-weight group would experience greater gains in strength. The machine-based training group experienced greater increase in machine bench press strength compared with the free-weight training group. This finding supports the idea of specificity, which refers to the concept that the greater the similarity training has to the actual performance test, the greater the probability of transfer (23). This training specificity is supported by one other study where training with free weights or using a Nautilus machine resulted in greater gains in strength when testing was performed on the training device (2). Also, 10 weeks of training with free-weight squat exercise was superior for increasing strength in the squat exercise compared with training with machine leg press (22). In contrast to the concept of specificity, our free-weight bench press, free-weight squat, and Smith machine squat increased in both the free weight and the machine training groups with no differences between the 2 groups. One previous study evaluating training with free-weight squats, the Nautilus Compound Leg Machine, or Universal Variable Resistance Maximum Overload Leg Press also found significant increases in all strength measures, which were similar across free weight or machine training groups (28). These findings do not support the concept of specificity, but rather, they show that there was good transfer of strength from one mode (i.e., free weights or machines) to the other. It is possible that the training level of our subjects contributed to a lack of specificity for training. Our subjects had on average over 2 years of resistance training experience and therefore may have had adequate previous experience training on different machines and with free weights. Individuals who are less accustomed to resistance training may experience greater specificity in their response to training (23).
Our results indicated that men and women responded in a similar fashion to the training program for increases in muscle thickness and strength (i.e., there were no sex × time interactions). This is in agreement with the early studies by Staron et al. who found similar increases in muscle fiber size (30) and strength (29) in response to training programs in women and men.
A surprising finding is that there was no difference in predicted 1RM or actual 1RM for squat or bench press between the Smith machine and free-weight modalities. One would expect a greater strength performance with the Smith machine because it offers greater stabilization. Previous studies comparing Smith machine and free-weight squat did not report 1RMs (1,25). In support of our bench press results, one other study found no difference for 1RM performance between Smith machine and free-weight bench press (24). They proposed that the linear motion of the Smith machine is unnatural and places the lifter at a biomechanical disadvantage, therefore, resulting in a lower lifting performance than expected. This may account for the similar 1RM between Smith machine and free-weight modalities.
The only group that experienced a significant acute increase in free testosterone was the men training with free weights. This result partially supports our hypothesis that the free-weight group would have a greater anabolic response to training. Recruitment of large amounts of muscle mass may be needed to elicit an acute free testosterone response (11). This greater acute anabolic hormone response for men exercising with free weights vs. machines is supported by one other study (26). In our study, the men training with machines experienced only a small nonsignificant increase in free testosterone. Although the acute workouts for the free-weight group and the machine group followed a protocol with similar training volume, the men training with machines in our study may not have received enough mechanical stress by training in the very stable environment of the Smith machine. Free-weight exercise requires more stabilization than Smith machine exercise as evidenced by substantially higher muscle recruitment, as assessed by electromyography (25). The stability and balance needed for the free-weight training session may have added the needed stress resulting in an acute free testosterone increase. The women, regardless of the training mode, did not experience any changes in acute free testosterone levels, similar to other studies (8,13). Overall, our salivary hormone results are similar to others who have assessed salivary hormone response to acute and chronic resistance training. Others have found that training with free weights (5,6) but not with programs that use mainly machines (10) results in an acute increase in salivary free testosterone in men but not women (19). Also, there is generally a lack of increase in salivary free testosterone with chronic training programs in men and women (20,27).
Although there was no acute change in free testosterone for women and men training with machines, they still had increases in biceps and quadriceps muscle thickness over the 8 weeks of training. This finding indicates that there may not be a direct causal relationship between increases in muscle mass and acute exercise-induced increases in free testosterone. This idea is supported by several recent studies that failed to find correlations between acute increases in anabolic hormone response and chronic changes in muscle mass during resistance training programs (16–18).
There are a number of limitations to the study. The machine program used a variety of machines, some of which had “cam” pulley systems and some which did not. The machines therefore would have differed biomechanically, and our results therefore are not specific to any one design of machine. Another limitation is that the training program was only 8 weeks in duration. A longer training program may be necessary to elicit differences in training responses to machines vs. free weights. In the current study, we assessed salivary hormones, but these may not always reflect serum hormone concentrations. For example, Cadore et al. (3) found that although salivary free cortisol concentrations correlated with serum free cortisol in response to an acute session of resistance training, there was no correlation between salivary and serum free testosterone.
In conclusion, results of this study show that significant increases in strength and biceps and quadriceps muscle thickness can be achieved by training with only free weights or only machines. Men training with free weights have an acute increase in free testosterone after an acute resistance training session, but this does not seem to translate into greater increases in muscle size or lean body mass when compared with resistance training with machines. These results suggest that muscle size and strength can be increased to the same degree with either training modality.
Frequently strength and conditioning practitioners promote the use of free weights over the use of machines to enhance muscle size and strength. Our study indicates that the mode of training (i.e., free weights or machines) does not determine the degree of muscle hypertrophy or strength increase. The results of this study indicate that muscle size and strength was increased similarly between a free-weight–only training group and a weight-machine–only training group during 8 weeks of training in subjects with about 2 years of previous training experience. This suggests that if the goal of the training program is to increase muscle size or strength, then either modality of training can be used effectively to accomplish this outcome. It should be emphasized that these results may differ if training programs are longer than 8 weeks and in subjects with less training experience. Our study also indicates that although there is an increase in free testosterone from an acute bout of free-weight resistance exercise, this may not translate into larger gains in lean body mass, muscle size, or strength when completing a chronic resistance training program.
This study was supported by a grant from the Saskatchewan Academy of Sports Medicine. The authors declare no conflicts of interest.
1. Behm DG, Anderson KG. Trunk muscle activity increases with unstable squat movements. Can J Appl Physiol 30: 33–45, 2005.
2. Boyer BT. A comparison of the effects of three strength training programs on women. J Appl Sport Sci Res 4: 88–94, 1990.
3. Cadore E, Lhullier F, Brentano M, et al. Correlations between serum and salivary hormonal concentrations in response to resistance exercise. J Sports Sci 26: 1067–1072, 2008.
4. Candow DG, Chilibeck PD. Differences in size, strength, and power of upper and lower body muscle groups in young and older men. J Gerontol A Biol Sci Med Sci 60: 148–156, 2005.
5. Crewther B, Cronin J, Keogh J, Cook C. The salivary free testosterone and free cortisol response to three loading schemes. J Strength Cond Res 22: 250–255, 2008.
6. Geisler S, Aussieker T, Paldauf S, et al. Salivary free testosterone and free cortisol concentrations after two different resistance training
exercises. J Sports Med Phys Fitness 59: 1030–1035, 2019.
7. Gibala MJ, MacDougall JD, Sale DG. The effects of tapering on strength performance in trained athletes. Int J Sports Med 15: 492–497, 1994.
8. Häkkinen K, Pakarinen A. Acute hormonal responses to heavy resistance exercise in men and women at different ages. Int J Sports Med 16: 507–513, 1995.
9. Hayes LD, Bickerstaff GF, Baker JS. Interactions of free cortisol, free testosterone, and resistance training
: Influence of circadian rhythms. Chronobiol Int 27: 675–705, 2010.
10. Kraemer WJ, Loebel C, Volek J, et al. The effect of heavy resistance exercise on the circadian rhythm of salivary free testosterone in men. Eur J Appl Physiol 84: 13–18, 2001.
11. Kraemer WJ, Ratamess NA. Hormonal responses and adaptations to resistance exercise and training. Sports Med 35: 339–361, 2005.
12. Kravitz L, Akalan C, Nowicki K, Kinzey SJ. Prediction of 1 repetition maximum in high-school power lifters. J Strength Cond Res 17: 167–172, 2003.
13. Linnamo V, Pakarinen A, Komi P, Kraemer W, Häkkinen K. Acute hormonal responses to submaximal and maximal heavy resistance and explosive exercises in men and women. J Strength Cond Res 19: 566–571, 2005.
14. Maddalozzo C, Snow C. High intensity resistance training
: Effects on bone in older men and women. Calcif Tissue Int 66: 399–404, 2000.
15. McCaw ST, Friday JJ. A comparison of muscle activity between a free weight and machine bench press. J Strength Cond Res 8: 259–264, 1994.
16. Mitchell CJ, Churchward-Venne TA, Bellamy L, et al. Muscular and systemic correlates of resistance training
-induced muscle hypertrophy. PLoS One 8: e78636, 2013.
17. Morton RW, Oikawa SY, Wavell CG, et al. Neither load nor systemic hormones
determine resistance training
-mediated hypertrophy or strength gains in resistance-trained young men. J Appl Physiol 121: 129–138, 2016.
18. Morton RW, Sato K, Gallaugher MPB, et al. Muscle androgen receptor content but not systemic hormones
is associated with resistance training
-induced skeletal muscle hypertrophy in healthy, young men. Front Physiol 9: 1373, 2018.
19. Nunes JA, Crewther BT, Ugrinowitsch C, et al. Salivary hormone and immune responses to three resistance exercise schemes in elite female athletes. J Strength Cond Res 25: 2322–2327, 2011.
20. Nunes JA, Crewther BT, Viveiros L, De Rose D, Aoki MS. Effects of resistance training
periodization on performance and salivary immune-endocrine responses of elite female basketball players. J Sports Med Phys Fitness 51: 676–682, 2011.
21. Rønnestad BR, Nygaard H, Raastad T. Physiological elevation of endogenous hormones
results in superior strength training adaptation. Eur J Appl Physiol 111: 2249–2259, 2011.
22. Rossi FE, Schoenfeld BJ, Ocetnik S, et al. Strength, body composition, and functional outcomes in the squat versus leg press exercises. J Sports Med Phys Fitness 58: 263–270, 2018.
23. Sale D, MacDougall D. Specificity in strength training: A review for the coach and athlete. Can J Appl Sport Sci 6: 87–92, 1981.
24. Schick EE, Coburn JW, Brown LE, et al. A comparison of muscle activation between a Smith machine and free weight bench press. J Strength Cond Res 24: 779–784, 2010.
25. Schwanbeck S, Chilibeck PD, Binsted G. A comparison of a free weight squat to a smith machine squat using electromyography. J Strength Cond Res 23: 2588–2591, 2009.
26. Shaner AA, Vingren JL, Hatfield DL, et al. The acute hormonal response to free weight and machine weight resistance exercise. J Strength Cond Res 28: 1032–1040, 2014.
27. Shariat A, Kargarfard M, Danaee M, Bahri Mohd Tamrin S. Intensive resistance exercise and circadian salivary free testosterone concentrations among young male recreational lifters. J Strength Cond Res 29: 151–158, 2015.
28. Silvester J, Stiggins C, McGown C, Bryce G. The effect of variable resistance and free-weight training programs on strength and vertical jump. J Strength Cond Res 3: 30–33, 1982.
29. Staron RS, Karapondo DL, Kraemer WJ, et al. Skeletal muscle adaptations during early phase of heavy-resistance training
in men and women. J Appl Physiol 76: 1247–1255, 1994.
30. Staron RS, Malicky ES, Leonardi MJ, et al. Muscle hypertrophy and fast fiber type conversions in heavy resistance-trained women. Eur J Appl Physiol Occup Physiol 60: 71–79, 1990.