Introduction
An area of strength training, which has been somewhat controversial, concerns the mode of exercise, i.e. free weights or machines. Often fitness instructors and particularly strength and conditioning coaches are bombarded with advertisements that expound upon the virtues of various types of resistance training equipment. Making appropriate selections of resistance training equipment can be paramount in producing maximum performance gains. The purpose of this review is to briefly examine the role of different modes of strength training in strength training programs. The following definitions will be used in for purposes of this discussion:
Free weights: barbells, dumbbells, associated benches, medicine balls body mass and augmented body mass (i.e. weighted vests and limb weights) - a freely moving body which does not inhibit the occurrence of normal force/acceleration patterns. Challenges the lifter to control, stabilize and direct a movement.
Machines: devices that apply resistance in a guided or restricted manner. Lesser challenge, compared to free weights, for stabilization, control and directed movement.
Strength Gains
Strength can be defined as "the ability to produce force" and has both a magnitude (0-100 %) and a direction. Two important questions will be considered in this discussion:
1) What is the magnitude of strength gains when using machines or free weights?
and most importantly
2) How transferable is the gain in strength to other performances?
Testing & Training Specificity: Both machines and free weights can produce strength gains. However, the magnitude of maximum strength gains as a result of resistance training largely depends upon the similarity between the strength tests and the training exercise. This aspect of mechanical specificity has been noted in both longitudinal studies (Abernathy and Jurimae 1996; Fry et al. 1992; Rasch and Morehouse 1957) and reviews of the literature (Behm 1995, Rutherford and Jones 1986; Sale 1988; Sale 1992; Stone et al. 2000).
Reasonable adherence to testing specificity can be very important in attempting to ascertain gains in maximum strength. Lack of adherence to testing specificity may explain why some studies failed to show differences between using free weights or machines. For example, in the studies by Saunders (1980) and Silvester et al. (1982) training was dynamic and testing was isometric which likely masks or reduces any gains in maximum strength and may reduce observable differences between groups (Wilson and Murphy 1996). One assumes that the rationale behind using a non-specific test for maximum strength is that such a test would not favor either mode of training. However, a device, which is supposedly non-specific usually, favors either the free weight or the machine. This results from the fact that in dynamic testing the device used for testing must be either a free weight or a machine. For example: in the study by Messier and Dill (1985) comparing Nautilus to free weight training, tests of leg strength were performed on a Cybex II semi-isokinetic leg extension device. This leg extension device is an "open kinetic chain exercise". The machine (Nautilus) group used leg extensions (open kinetic chain exercises) in training, however, the free weight group used squats ("closed kinetic chain") and no leg extensions in training. Thus the machine group may have had an advantage in testing as part of their training was mechanically more similar to the testing mode.
Even though differences between groups using different devices for training can be masked or reduced by using a non-specific measurement, some transfer (i.e. carry-over) between devices may be expected (Knapik et al. 1983, Wilson and Murphy 1996). However, it is likely that the effect size (the degree of average change between groups over time taking into account individual subject change) must be quite large in order to show any differences. Interestingly, most of the available peer-reviewed published data indicates that maximum strength (1RM) gained as a result of free weight training can carry-over to machine testing better than the converse (Boyer et al. 1990, Stone et al. 1979, Wathen and Shutes 1982). The idea of greater carryover in maximum strength testing (1 RM) is supported by two unpublished observations from our laboratories (Jesse et al. 1988; Brindell 1999). The exact reasons for greater carry-over from free weights to machines is unknown. However, it is reasonable to speculate that skilled movements involving a relatively high degree of motor coordination may have greater transferability compared to movements requiring less motor coordination.
There are several problems with some of these studies which preclude any definitive conclusions concerning comparison between different modes of training. These problems range from lack of specific testing (see above), using different protocols and too few subjects. Perhaps the two most important problems are, 1) study length, most studies last only a few weeks (i.e. 5 - 12 wks), unfortunately there are no long term (i.e. years) studies and 2) training status (Stone et al. 2000). However, based on the available information, there is an indication that 1 RM measures of strength carry over from free weights to machines better then machines carryover to free weights.
Transfer of Training Effect: Careful observation indicates that skillful free weight movements can augment motor control and general coordinative abilities to a greater degree than machines (Drabik 1996; Harre, 1982, Siff and Verkhoshansky 1998). These "coordinative abilities" can include:
| • Orientation and differentiation |
| • Reactive ability |
| • Rhythm and balance |
| • Combinatory abilities |
Additionally, strength and conditioning coaches have generally advocated the use of free weights in the training of various types of athletes. Much of the reasoning for using free weights as a primary mode of training results from their observation that free weights "carry-over" to performance better then machines (Roundtable 1993).
These observations suggest that the major contributing factor to the superiority of free weights compared to machines is the ability to mimic and overload most athletic (and daily task) movements. Thus, a priority for training should be to use exercises specific to competitive movements. In order to facilitate carry-over from training exercises to performance appropriate force characteristics such as magnitude, rates of force development, power and speed must be properly overloaded. Additionally movement pattern specificity must be considered and properly simulated by the training exercise (Stone et al. 2000). Thus, a greater degree of mechanical specificity afforded by free weights could facilitate a greater transfer of training effect (Stone et al. 2000). Using the vertical jump as an example:
The vertical jump is often chosen as an indicator of explosive athletic performance because: 1) it is easy to measure, 2) the vertical jump is a primary component of many sports (i.e. basketball, volleyball etc. etc.), 3) there are reasonable associations/correlations between the vertical jump and performance in other "explosive" exercises, for example sprinters jump higher and sprint faster than distance runners (Hollings and Robson 1991), 4) the VJ (or its components including velocity and power output) have been associated with the level of performance ability in numerous specific sports (Anderson et al. 1990; Barker et al. 1993; Stone et al. 1980; Thissen-Milder and Mayhew, 1991).
With few exceptions (Wathen and Shutes 1982) free weights have consistently produced superior results in vertical jump gains across short term training periods (Augustsson et al. 1998, Bauer et al. 1990, Silvester et al. 1982, Stone et al. 1979, Wathen, 1980).
As previously pointed out, specificity dictates that a number of kinetic and kinematic parameters must be appropriately overloaded to stimulate gains in performance. One of the most studied and contemplated performance aspects of specificity is the vertical jump (VJ) and its relationship to weightlifting movements (snatch, clean & jerk and derivatives) and the training practices of weightlifters. Indeed improved weightlifting performance as a result of training has been associated with increased vertical jump height and associated power output among novices (Stone et al. 1980). Furthermore, weightlifters have been shown to have superior weighted and un-weighted vertical jump heights and power outputs compared to other athletes (McBride et al. 1999). Part of the reason for these superior performance characteristics is likely related to the mode and methods used by weightlifters in training. Although adaptations to training are always multi-factorial, one likely contributing factor is the degree of associated mechanical specificity which has been observed between weightlifting movements (i.e. snatch, clean & jerk and derivatives) and the vertical jump (Canavan et al. 1996; Garhammer 1981). These factors include a combination of high power output, high rates of force development and movement patterns, which cannot be easily duplicated by machine use.
Other assumptions have been made concerning the use of free weights and machines, which cannot be supported when examined closely. For example, it is often assumed that throwing motions requiring twisting (trunk rotation) cannot be made and trained appropriately with free weights and that machines are necessary. However, this idea may be more related to lack of experience with free weight training rather than the actual mechanics of free weights or machines. First it should be remembered that most throwing movements are made in a standing or upright position. For many years throwers have simulated this upright positions using weighted balls and implements; additionally walking twists and weighted hammer thrower exercises have been used successfully to overload upright trunk rotation and throwing motions. Furthermore, with the use of benches or pommel horses a variety of positional exercises using both weights and balls can stress trunk rotation from a variety of angles, which cannot be attained with most machines.
While most physical activities can be simulated and appropriately trained using free weights, possible exceptions are some aspects of swimming in which motion is generated in a supine or prone position largely through propulsion by the upper body. In this case it may be advantageous to use a "swim bench" to simulate and overload stroke mechanics.
Relative advantages and disadvantages of free weights and machines:
- FW are typically cheaper and thus have greater cost effectiveness. For example: Well made olympic barbells can be purchased in the United States, Great Britain and most European countries for well under $1000.00 for a 400 lb. set. If the cost of outfitting a major training facility with typical machines (each one performs essentially one function) were used to purchase barbells, dumbbells and associated benches, then this would allow substantially more people to train simultaneously at the same cost (or less). Additionally machines generally require more maintenance than do free weights.
- Most machines have limited adaptability - that is the machine will only allow performance of the exercise (with few variations) for which it is designed. This is not a problem with free weights in which exercises can be created to fit the activity (i.e. greater degree of mechanical specificity).
- Although some manufacturers have attempted to improve adjustment factors most available machines do not have sufficient adjustment capability to be able to fit all sizes or populations. Even cursory observation of athletes or non-athletes reveals differences in height, weight, limb length etc. which will affect the way in which many, particularly variable resistance, devices are able to effectively apply resistance. For example most machines are made to fit adults not children. An advantage of free weights is that "one size fits all". In this context "variable resistance" machines do not adequately apply resistance in relation to human strength curves, particularly on an individual basis (Cable and Zebas 1999). Variable resistance machines attempt to apply resistance, which matches human strength curves. However, these devices do not necessarily accurately match an average human strength curve and do not relate to individual strength curves which depend upon individual biomechanical factors such as limb lengths.
- Metabolic aspects of exercise and training can be very important to both athletes and non-athletes. Large muscle mass exercises require more energy than smaller muscle mass exercises. Body mass and body composition can be greatly influenced by total energy expenditure, thus large muscle mass exercises are more likely to result in positive body composition and metabolic adaptations (for review see Stone et al. 1991). A greater variety of large muscle mass exercises can be performed with free weights as compared to typical machines. In this context, free weights can be advantageous in terms of time-efficient training sessions. One large muscle mass, multi-joint exercise such as the squat press can exercise as many muscle groups as 4-8 small isolated or small muscle mass exercises, thus saving time.
- The development of training protocols in which the exercises have a high degree of mechanical specificity (with appropriate training design) is the major advantage of free weights. The use of these protocols is particularly important for developing speed and power. Mechanical specificity includes force characteristics (i.e. magnitude of force, rate of force development, velocity, power) as well as movement patterns. There is little doubt that free weights can satisfy these aspects of specificity better than typical machines. (See Stone et al. 2000 and Siff and Verkoshansky 1998 for more detailed discussions).
- Space limitations can be a disadvantage for free weights (and most machines). For example, space can be limited in confined quarters such as aboard ships, submarines, space vehicles etc. In many cases specifically constructed machines, sometimes using springs or elastic bands can take up less space.
- In some specific small muscle mass exercises in which joint segments "close in on themselves" such as arm curls, some machines may offer resistance through a greater range of motion compared to any one free weight exercise. One example would be cable crossovers compared to typical flys when lying on a bench. This may be an advantage in hypertrophy development with some machines.
- It is often argued that machines are safer than free weights. However, there is no evidence for this assumption (Requa et al 1993). In the author’s experience, there are as many or more injuries occurring with machines as free weights. Often these injuries result from poor technique or poor technique coupled with poor training programs regardless of the mode of exercise.
Use of Free Weights with Non-Athletic Populations
This review has been primarily concerned with strength power training adaptations for competitive athletes. However, the coach (and other fitness specialists) may be called upon to advise or assist with the training of recreational athletes or non-athletic groups.
So, it is necessary to have some knowledge of the potential use of various modes of resistance training for these groups It has been assumed that certain non-athletic populations, particularly populations comprising the elderly or certain disease states, such as arthritis, cannot use free weights due to physical or psychological limitations. This assumption is largely based on real or perceived limitations such as: 1) weight-bearing inability (either whole body or specific segments) as a result of pain or weakness, 2) psychological factors such as the free weights are intimidating and 3) free weights require more "technique training" and supervision. However, this is an assumption that has not been adequately tested. In fact evidence suggests that using free weights can be a safe an effective method of enhancing performance in non-athletic populations, including aging populations in which the frequency and severity of degenerative diseases would be increasing. For example, among sedentary men ranging in age from 30 - 60 years, training programmes employing primarily free weights have resulted in a number of beneficial alterations including increased maximum strength, power and beneficial alterations in blood lipids (Johnson et al. 1982, Blessing et al. 1987). More recently, Brill et al. (1998) successfully used a free weight program with an elderly population (73 - 91 years) in promoting beneficial adaptations in several functional performance measures (such as balance, stair climbing, etc.). No adverse effects were noted.
The important aspect to consider is "primary" exercises. Training exercises should be primarily carried out with free weights for the same reasons that athletic populations should used them. There is no reason to believe that the superior "transfer of training effect" which can be realized from free weights would not be effective in improving daily tasks such as lifting, carrying, shoveling etc. In this respect it should be noted that "free weights" do not have to take the traditional form of barbells and dumbbells; rather weighted vests and limb weights can be used to advantage among some groups, such as frail or elderly individuals. By using this form of free weights daily activities can be directly overloaded through augmented body or limb mass movements. For example, rising from a chair can be trained using a weighted vest. While a few machine exercises may be advantageous most of the exercises should be performed with free weights for all populations. Exceptions are not usually population oriented but rather situation oriented, for example, where space may be at a premium (i.e. a submarine - crews have used elastic bands which take up less space then either free weights or most machines). Indeed, the likelihood of not being able to perform a particular exercise may be more a function of individual physical and psychological characteristics, which may be coupled with specific disease states or injuries, rather than characteristics of a population. Competent strength training personnel can easily recognize these individual problems and programme adaptations can be made accordingly. In this context both authors have a background of training not only competitive athletes, but also of working with or supervising strength training programmes which have included recreational athletes, disabled athletes, middle aged and elderly groups. While there are individuals with problems, which preclude the use of certain free weight exercises, it is our opinion, that most individuals can safely and effectively use training programmes primarily based around free weights.
Summary
While a number of factors would seem to indicate that free weights can produce superior results, the major factor appears to deal with mechanical specificity. From the standpoint that specificity of exercise and training can result in superior adaptations free weight training should produce a greater carry-over to other performance aspects of sport.
|
