Journal Logo


Training Loads for the Development of Lower Body Muscular Power During Squatting Movements

Hansen, Keir MHSc1,3; Cronin, John PhD2,3

Author Information
Strength and Conditioning Journal: June 2009 - Volume 31 - Issue 3 - p 17-33
doi: 10.1519/SSC.0b013e3181957065
  • Free



A variety of loading schemes have been used in research to examine the most effective means of developing muscular power. Both heavy load-low velocity training (24,32) and light load-high velocity training (13,14,16,18,23,34) have been extensively researched to establish the most effective means of developing muscular power and improve muscular performance. Given that power is the product of force and velocity, it is possible that training at a heavy load will increase force output and training at a light load improve velocity. Therefore, either approach may improve the power output of musculature as long as there is not a concomitant decrease in force or velocity (depending on the training emphasis). It has been widely suggested in the literature that perhaps the load that maximizes mechanical power output should be used for optimal improvement of power output (3,19,34). This may provide the ideal balance between force production and velocity of movement during power training.

Given the debate as to the optimal loading for power development, this work will review the literature investigating the effect of different training loads on force, velocity and power qualities, and sports-specific measures in the lower body after lower-body strength training interventions. For the purposes of this review, training studies have been categorized as heavy load (>70% of 1RM training load, n = 8), moderate load (20-70% of 1RM, n = 6), and light load (body weight [BW] only, n = 5) training and mixed-load training (a combination of 2 or more of the aforementioned loads n = 5). To disentangle the effect of these various training loads, each section discusses the magnitude of change in maximum strength, force, velocity, power, and sports-specific performance, by calculating and comparing percent changes and effect sizes (ES). The ES allows us to compare the magnitude of the treatment (strength program) on variables between studies. We describe the effects as “trivial,” “small,” “moderate,” and “large” based on the description of effects for untrained, recreationally trained, and highly trained athletes (31). Such classification means that effect sizes are not described in a uniform manner throughout the different populations (Table 1).

Table 1
Table 1:
Interpretation of effect sizes relative to training status as described by Rhea (31)

Seven databases were searched for power training studies, including PubMed, MEDLINE, SPORTdiscus, Web of Science, Proquest, Meditext, and Education Full Text. The selection method of the studies gathered during the literature search involved one reviewer performing the selection of studies in 2 consecutive screening phases. The first phase consisted of selecting articles based on the title and abstract. The second phase involved applying the selection criteria to the full-text articles. Studies were chosen if they fulfilled the following 6 selection criteria: (a) the study used a training method that corresponded to one of the loading schemes previously described; (b) the study detailed the training program and used the squat, jump squat (JS), or unloaded jumps as the primary training and testing movement pattern; (c) the outcome measures of interest were clearly detailed; (d) studies that did not provide group means and standard deviations before and after training were excluded as comparing percent changes (pre- to post-training) and effect sizes were the primary means of analysis; (e) studies were published between 1985 and 2008; and (f) studies had to have been written in the English language and must have been published as a full-text article in a peer-review journal. Abstract-only publications were not included.


The age of subjects ranged from 18 to 61 years, and only 2 studies included female subjects. However, the training status of subjects varied considerably. According to the classification system of Rhea (31), 8 studies had an untrained subject population (<1 year of resistance training experience), 13 studies had a recreationally trained population (1-5 years of resistance training experience), and none had a highly trained population (>5 years of resistance training experience). Given this fact the findings of training studies into power training across the loading spectrum must be applied to highly trained populations with great caution.

The design of the training interventions is obviously a key factor in the training adaptations produced during training studies. The variation within the squat/JS power training research reviewed is disparate, as can be observed from the Tables 2 to 5. In terms of training volume, if the most simple calculation of total training volume is used (volume = sets × reps × load), it is clearly evident that there is a large disparity in training volume both within studies investigating the effects of a particular load, and between training loads. This is further confounded by the inconsistency in selection of training frequency, number and choice of movement patterns, training tempo and rest periods. These issues are particularly evident when examining studies using bodyweight (plyometric) training techniques. Studies use a variety of movement patterns that include single-leg and double-leg movements, vertical and horizontal movements, and depth jumps, making the quantification and comparison of the overload provided almost impossible. The reader needs to be cognizant of these limitations and the comparison within and between studies must be undertaken with caution.

Table 2
Table 2:
Effects of heavy load lower body (>70% of 1RM) squat and jump squat training on muscular power and sports-specific measures
Table 2
Table 2:
Table 2
Table 2:
Table 3
Table 3:
Effects of moderate load lower body (<70% of 1RM) squat and jump squat training on muscular power and sports-specific measures
Table 3
Table 3:
Table 4
Table 4:
Effects of light load lower body (plyometric) squat and jump squat training on muscular power and sports-specific measures
Table 5
Table 5:
Effects of mixed load lower body squat and jump squat training on muscular power and sports-specific measures
Table 5
Table 5:


In this section, we review the literature investigating the effects of heavy load training (>70% 1RM) squat/JS training on force, velocity and power output, as well as performance in sports specific tasks.


Maximum strength as measured by squat 1RM (see Table 2) has been shown to increase with heavy-load training (16,23,25,26). Reported percent changes in 1RM range from 6.1% (26) to 21.9% (35), which represents effect sizes from 0.17 to 1.64, the latter of which can be considered a moderate training effect for the untrained population investigated. The discrepancy in training changes in maximum strength amongst training studies investigating heavy loads can be explained largely by the inconsistencies in training prescription and the differences in subject populations as already discussed. For example, Young and Bilby (35) had untrained subjects perform 4 sets at 8-12 RM 3 times per week for 7½ weeks at a slow tempo, which resulted in a 21.9% increase in back squat 1RM. Harris and colleagues (16) who investigated recreationally trained subjects prescribed 1 set of 6-8 repetitions 3 times a week for 8 weeks at 80% of 1RM, resulting in a 9.8% shift in squat 1RM. Although the methods used for quantifying load was different between these 2 studies, it would seem clear that the study by Young and Bilby (35) involved a greater training volume and accordingly greater strength increases would be expected. A number of studies used untrained subjects and consequently reported large shifts in various training parameters. For example, Young and Bilby (35) investigated an untrained population and reported a 19.9% (ES = 1.64, moderate) increase in squat 1RM. However, Harris and colleagues (16) used a population that would be classed as recreationally trained and reported only a 9.8% (ES = 1.86, moderate) increase in squat 1RM.

A number of studies using heavy loading parameters reported changes in force production capabilities (peak force, average force, and rate of force development [RFD]) during both isometric and dynamic tasks. Changes in these parameters also varied greatly across training studies. Young and Bilby (35) reported a 45.5% (ES = 0.83, small) increase in RFD during a vertical jump after 7½ weeks of training in untrained athletes and Wilson and colleagues (34) reported a 10% increase (ES = 0.21, trivial) in isometric maximum RFD after 10 weeks of training in subjects with 1 year of resistance training experience (see Table 2). Peak force (PF) also has been measured using a variety of means before and after training, including isometric PF and JS PF at a variety of loads. However, as with maximum strength changes, the comparison of results produced by training studies is difficult as a result of the large variation in training prescription and subject populations. This difficulty is further confounded by the variety of movement patterns and testing loads used during the measurement of force parameters in the research reviewed.

Jump squat PF data have indicated some load adaptions after heavy load training. For example, McBride and colleagues (23) reported that after a training period performing JS at a load of 80% of 1RM, subjects significantly increased (p < 0.05) PF during JS at 55% and 80% of 1RM. This study reported PF increases of 4.84% (ES = 1.09, moderate), 7.37% (ES = 1.67, large), and 7.18% (ES = 1.45, moderate) for 30% 1RM, 55% 1RM, and 80% 1RM testing loads, respectively. Similar findings were reported by Jones and colleagues (18), who reported a 2.2% (ES = 0.22, trivial) increase and a 6.9% (ES = 0.50, small) increase in peak force during JS at testing loads of 30% and 55% of 1RM, respectively. These studies, which both used recreationally trained subjects, tend to suggest that a load-specific training effect is evident in peak force production, with the greatest percent changes in peak force production and effect sizes occurring at testing loads closest to the training loads.


The studies of McBride and colleagues (23) and Jones and colleagues (18) are the only studies to have reported changes in velocity of a loaded movement after JS training (see Table 2). Both studies reported an increase (% and ES) in JS peak velocity (PV) at most tested loads (see Table 2). The exceptions were the 30-kg JS in the study of McBride and colleagues, which resulted in a 0.54% decrease and the BW squat jump in the study of Jones and colleagues, where a large 9% decrease was reported. When effect sizes are examined, none were moderate to large, with the greatest being an effect size of 0.47 (small) reported by Jones and colleagues for a JS at 30% of 1RM. These data would suggest that the effect of heavy load training, even when the intent is to move the load as rapidly as possible, does not elicit significant increases in velocity of movement even at the prescribed training load.


If this is the case, that high load training can illicit changes in force production but not velocity of movement, then one would anticipate a shift in power performance based on an increase in force capability (so long as velocity of movement was not negatively affected). Power changes after high load training have been extensively reported in the training literature during the squat and JS movement. McBride and colleagues (23) reported that peak power (PP) increased with a moderate or large effect size after training at 80% of 1RM at both the heavier testing loads (55% and 80% 1RM). These loads corresponded with those that showed significant improvements in peak force production. However, unlike McBride and colleagues, the research of Jones and colleagues (18) reported greater improvements in PP during a JS at 30% of 1RM than at 50% of 1RM (5% versus 2.9%); however, the effect size (ES = 0.27 and ES = 0.33 for 30% and 50% 1RM, respectively) at both loads would be considered trivial.


Many studies have included measures of sports-specific tasks such as jumping movements and sprinting over a variety of distances in their investigation of adaptation to training (see Table 2). Most who have used the vertical jump (VJ) have reported that heavy load training has a positive effect on performance. Reported percent changes range from 2.3% (16) to 7.9% (35). No studies in which the authors used jumping movements as a sports-specific assessment after heavy load training showed moderate or large effect sizes (ES = −0.02 to 0.43). The study of Neils and colleagues (26) was the only one that reported decreases in jumping performance after heavy load training, reporting decreases in both squat jump (SJ) and countermovement jump (CMJ) performance.

In terms of the effects of heavy load training on sprint performance, only Murphy and Wilson (25) and Delecluse (10) reported a decrease in sprint times of −0.22% and −0.24%, respectively (improved performance), and neither reported this change as being statistically significant. Many of the studies reviewed (11,18,23) actually reported an increase in sprint times (decreased performance) after heavy load training. These negative performance changes ranged from a 1.07% decrease in 10-meter acceleration performance reported by Delecluse (10) to a 6.1% (ES = 2.33, large) and 4.89% (ES = 3.00, large) decrease in 5-meter and 10-meter performance reported by McBride and colleagues (23). Therefore, it seems that even with positive adaptations in terms of maximum strength and selected force and power variables, heavy load training does not have a positive effect on power and speed related sports specific tasks.


In this section, we review the literature investigating the effects of moderate load training (20-70% 1RM) squat and JS training on force, velocity and power output, as well as performance in sports-specific tasks. Typically, these loads are selected in training to maximize power output. Kaneko and colleagues (19) reported that a 30% 1RM load maximized mechanical power output and maximized power adaptations after training. However, JS research has shown that the load that maximizes mean and peak power output may be dependent on the athlete's training age, exercise technique, equipment, and data analysis calculations (3,6-8,12), which has resulted in some inconsistency in determining what this load is. Nonetheless, a spectrum of loads (from 30% to 60% of 1RM) has been investigated to examine the effect of moderate loads on athletic performance.


Increases in maximum squat strength (1RM) resulting from moderate load training range from 3.6% (16) to 14.1% (22), both of which represent a small effect size (ES = 0.45 and 0.64, respectively). McBride and colleagues (23) reported only an 8.2% increase in 1RM squat but this equated to a moderate effect size (ES = 1.22). Greater increases in squat strength were reported by Harris and colleagues (16). The 15.5% increase however, was measured during the quarter squat. Again, much of the difference in the results between these 2 studies may be explained by program design, with some notable differences in the training intervention. Possibly the most important of these differences was that the study of McBride and colleagues used a ballistic movement (JS) whereas Harris and colleagues used a nonballistic traditional squat and quarter squat. Both of these studies investigated a population with some resistance training experience, which indicated that depending on training prescription, moderate load (∼20-30% of 1RM) ballistic training can elicit increases in maximum strength.

Two studies reviewed investigated the effect of moderate load training on RFD. Wilson and colleagues (34) reported a 10.8% (ES = 0.25, trivial) decrease in isometric maximum RFD after training, whereas Kyrolainen and colleagues (20) reported 17.9% increase (ES = 0.79, small) in knee extensor maximum RFD after training. The different results reported can again be explained by differences in training prescription and testing methodology. The training program administered by Kyrolainen and colleagues used JS at a variety of loads (30-60% 1RM), whereas Wilson and colleagues used only a 30% training load. Differences also may be explained by the testing methodology, as Wilson and colleagues (34) performed isometric testing using the squat movement, whereas Kyrolainen and colleagues (20) performed an isolated knee extension movement. It seems that this area requires further research, particularly relating to the assessment of RFD during compound iso-inertial movements. Only then can the effect of different loading schemes and training prescription be assessed and applied to strength and conditioning practice.

McBride and colleagues (23) reported moderate-to-large effect sizes in PF enhancement at all testing loads after 8 weeks of JS training at 30% 1RM. Interestingly, the weakest training effect occurred closest to the training load (30% 1RM), with greater training effects observed at the heavier testing loads (6.0% and 5.6% change for 55% 1RM and 80% 1RM, respectively). However, Jones and colleagues (18), reported the greatest change in PF at 50% 1RM, which was closest to the training load (40-60% 1RM). However, none of the force changes reported by Jones and colleagues were classified as large effect sizes.

Again, the ballistic nature of the training prescribed by McBride and colleagues resulted in greater force adaptations, largely one would speculate, as the result of the adjusted acceleration profile of ballistic movements performed at light to moderate loads (9,29). Research has shown that during ballistic movements at moderate to light loads greater forces are produced later in the movement due to the load being accelerated for longer periods when compared to traditional movements (where deceleration starts relatively early in the movement) (9,29). Thus, it seems that to elicit substantial changes in PF at moderate-to-light loads, movements must be performed in a ballistic manner.


The principle of specificity would suggest that moderate-to-low load training performed at high velocity may be the best way to elicit increases in movement velocity. Indeed, McBride and colleagues (23) found a significant (p < 0.05) increase in PV at all 3 JS testing loads (30%, 55%, and 80% of 1RM). However, as can be observed from Table 3, these significant changes only resulted in small or trivial effect sizes at all 3 testing loads tested. Jones and colleagues (18) also reported only trivial to small effect sizes despite a large percent change (12.4%) during a 30% 1RM JS. These data indicate that the velocity component of power may be the most difficult to shift in training. The percent change data and effect sizes for PV after moderate load ballistic training tend to be greater than those resulting from heavy load training (Tables 2 and 3), and, accordingly, the moderate load method may be the preferred option for improving velocity of movement. However, because there were no moderate or large effect sizes for velocity values, it is likely that it is very difficult to elicit large changes in velocity values during training. Alternatively, it may be that current assessment procedures may not be sensitive enough to monitor changes in PV as a training outcome. Nonetheless, it seems that even when training with moderate loads, change in force (using current assessment procedures) is greater than change in velocity following a training intervention.


PP has been measured using a number of methods and a variety of loads, resulting in percent changes ranging from 2.4% (ES = 0.57, small) to 16.4% (ES = 2.81, large). Interestingly, the greatest percent change in PP occurred in the study of McBride and colleagues at the 80% 1RM testing load. In this study, the percent change and the ES increased as testing load increased. These findings seem to oppose those proponents of load-velocity specific adaptation, the moderate loads used in this study resulting in a crossover in power adaptation from the lighter training loads to heavier loads. However, this was not evident in the study of Jones and colleagues, who reported the greatest percent change at the 30% 1RM testing load. In general, percent changes and effect sizes of PP measures were greater following moderate load training compared to heavy load training. When ballistic movements were utilized in training, a shift in both PF and PV were evident resulting in a greater overall increase in PP.


A variety of sports-specific tasks have been used to measure performance changes following moderate load training. Wilson and colleagues (34) reported 30% 1RM to be the load that developed all-round athletic performance most efficiently. They reported increases in CMJ (ES = 1.03, moderate, 16.8%), SJ (ES = 1.02, moderate, 14.8%), and decreases in 30-meter sprint times (ES = −0.17, trivial −1.1%), which exceeded those resulting from both high load and plyometric training. Sprint times decreased in 2 of 3 distances (−1.6% and −0.9% at 10 and 20 meters, respectively) investigated by McBride and colleagues (23), in contrast to increases in times after high load training, although at both loads changes resulted in either trivial or small effect sizes. Accordingly, the literature (see Table 3) remains far from conclusive in terms of the ability of the adaptations induced from moderate load training to transfer to improvements in performance of sports-specific tasks.


In this section, we review the literature investigating the effects of light load training BW squat/JS training on force, velocity and power output, as well as sports-specific assessments. The use of BW jumping movements to develop muscular power is commonly termed plyometric training (5). In most cases, plyometric training involves the coupling of eccentric and concentric muscle actions, to develop the athlete's ability to use eccentric forces via the stretch shorten cycle (SSC) (2,4,15). Research into lower-body plyometric training methods has primarily focused on the ability of plyometric training to induce improvements in jump and sprint performance. Nonetheless, in the context of discussing the effect of load on power performance a brief discussion of these methods is pertinent.


Only one reviewed study investigated the effect of plyometric training on maximum strength performance. Fatouros and colleagues (13) reported a 12.4% (ES = 2.6, large) increase in squat 1RM after training. Given that the subjects in this study (13) were untrained, the increase in maximum strength with the addition of low load ballistic training was not surprising. It can be concluded that the current literature is inconclusive in terms of the ability of plyometric training, on its own, to shift maximum strength in the lower limb in subjects with any level of training experience.

There is a dearth of research that has examined the changes in force and velocity profiles across a spectrum of loads following plyometric training programs. Wilson and colleagues (34) examined isometric maximum RFD and isometric PF in the squat movement after 10 weeks of depth jump training and reported 11.5% (ES 0.26, trivial) and 0.7% (ES = 0.02, trivial) shifts, respectively. It is worthy of note that the use of an isometric test to examine training adaptation after a dynamic training intervention is not ideal; indeed, the lack of specificity of such assessment practices has been highlighted in the literature (1). Testing procedures assessing force qualities using ballistic movements such as jumps and JS after this type of training intervention may be more appropriate.


A number of studies have investigated power output during the vertical jump (13,21,30). Fatouros and colleagues (13) reported a 25.6% (ES = 1.7, moderate) increase in power output during a vertical jump after a 12-week plyometric training intervention with untrained male subjects. This program used a variety of movement patterns and managed training load through the number of foot contacts per session (these ranged from 80 up to 220 contacts per session). However, although this study resulted in a moderate training effect, it again involved the application of a relatively intense training stimulus to untrained athletes with a large window for adaptation, and accordingly the magnitude of the power improvements is unlikely to be the same in more well trained populations.

Luebbers and colleagues (21) reported a very small post-training increase in vertical jump power (0.31 % change, ES = 0.05). In comparison with the previous study, the subjects trained for only 7 weeks (compared with 12), used physically active subjects and implemented lower training volume in each session. Given these facts, it is not surprising that there was less improvement. These results suggest that in active and trained individuals' volume and duration of training must be carefully planned to elicit positive power adaptation. Holcomb and co-workers (17) examined changes in PP during the CMJ and SJ in 2 training groups (CMJ- and DJ-trained groups), resulting in improvements that produced moderate-to-small effect sizes (% change = 2.5% − 7.4%, ES = 0.12-0.60). There were greater improvements (% change) reported after DJ training than CMJ training, but these improvements were not statistically significant.


Again, the most common measures of sports-specific performance in the training literature were jumping and sprinting tasks. Jumping tasks including the VJ, SJ, and CMJ, resulted in posttraining changes ranging from a decrease of 0.31% (ES = −0.03, trivial) reported by Luebbers and colleagues (21) in the VJ to an increase of 13.6% (ES = 0.56, small) in the SJ reported by Gehri and colleagues (15). Fatouros and colleagues (13) reported a slightly smaller 11.3% increase in vertical jump height (ES = 2.5), which was the only large effect size reported for VJ performance among the studies reviewed, with the results classified as either moderate or small. Again, the variation in results reported reflects the disparity in the subject populations used and the design of the training interventions. For example, when comparing the studies of Luebbers and colleagues (21) and Gehri and colleagues (15), although the subject populations were similar, the training volume and exercise selection were very different. In the study of Gehri and colleagues (15), the training intervention included only multiple CMJs, whereas the study of Luebbers and co-workers included a variety of movements, which amounted to a greater total training volume.

Results with regard to sprint performance after training also were inconclusive. Wilson and colleagues (34) reported only a 0.2% (ES = −0.03, trivial) improvement in a 30-meter sprint after training. The sports-specific task affected the most by plyometric training was the Margaria stair climb test used by Luebbers and colleagues (21), who reported a 6.3% (ES = 0.40, trivial) improvement in performance after training. Therefore, the results of the studies reviewed make conclusions as to the efficacy of plyometric training in improving functional performance measures difficult. Once again this is confounded by the variation of training interventions and subject populations investigated.


Given the research discussed thus far it may be that the use of mixed load training offers the “best of both worlds” in terms of having the ability to develop both high movement forces and high movement velocities. Mixed load training for lower body power development has been used in a number of forms. These include the use of heavy, moderate and light loads within a given training session (28), alternating training loads between training sessions (27), and complex training, which involves super setting heavy and moderate or light loads during training (21). Intuitively, these training systems are appealing because they offer the opportunity for training to be done across the force-velocity-power spectrum. Nonetheless, despite the popularity of the squat and JS movements in training practice there is limited research investigating mixed load training in this movement pattern. In this section we review the literature investigating the effects of mixed load training squat/JS training on force, velocity and power output as well as sports specific performance.


There was a large range in maximum strength (squat 1RM) among the mixed load training studies reviewed. These ranged from an increase in squat 1RM of 1% (ES = 0.1, trivial) reported by Newton and colleagues. (28) to 47.8% (ES = 3.69, large) reported by Tricoli (33). Tricoli and colleagues and Lyttle colleagues (22), who reported the second highest increase in 1RM strength 12.7% increase (ES = 0.8, small-moderate) both prescribed a training program by using a combination of maximum strength training and depth jumping. However, Newton and colleagues (28) used mixed loads within a single session (Table 5).

Newton and colleagues (28) reported a 11.3%, 5.4%, and 5.4% increase in JS peak force for BW, BW + 20 kg, and BW + 40 kg, respectively (raw data were not provided to calculate effect sizes). These changes during squat jumps represented significant changes (p < 0.05) in the JS group as compared with a control group that performed traditional high load resistance training only. However, Newton and colleagues (27) reported changes in peak force during JS at a variety of loads ranging from 4% to 29% (see Table 5) but reported mean data only for some variables, so calculation of effect sizes for changes in force and power variables was not possible. These researchers also reported a 23% (ES = 1.6, moderate) and 26% (ES = 0.6, small) change in isometric squat peak force in younger and older men respectively following mixed load JS training.


Changes in peak power following mixed load training ranged from 2.6% (ES = 0.7, small) reported by Harris and colleagues (16) to a 36% increase reported by Newton and colleagues (27) after mixed load JS training with untrained older men. The research of Newton and colleagues indicated that greater increases in JS peak power occurred at higher training loads (see Table 5) after mixed load training, although, this may be a result of the untrained population having a low baseline power output at the higher testing loads. It has previously been reported that athletes with a strength training history my produce greater power outputs at greater loads (3). Accordingly, it is very difficult to make definitive conclusions as to the effect of mixed load training on power output for elite populations from the research currently available.


Changes in jump performance after mixed load training ranged from 2.9% (ES = 60.0, large) increase in vertical jump reported by Harris and colleagues (16) to a 14.2% (ES = 0.7, small) increase in squat jump height reported by Lyttle and colleagues (22). The research of Harris and colleagues (16) reported a very large effect size. However, this large effect size was largely the result of a very small pretraining standard deviation in vertical jump height as the actual percent change in jump performance post training was very small (pretraining mean VJ = 62.2 cm, SD = 0.03 cm, posttraining mean VJ = 64.0 cm). The difficulty in comparing mixed load studies is highlighted in the comparison of these 2 studies. Lyttle and colleagues (22) used high load squat training, combined with plyometric depth jumping in the same session, Harris and colleagues used training loads from 30-80% of 1RM on different training days. Further difficulty in comparing programs results from a diverse range of assessment techniques, with a number of different jumping methodologies used.

With regards to sprint performance, similar to other training loads, results were inconclusive. For example, Tricoli and colleagues (33) reported a 2.7% (ES = 0.47, small) and 0.8% (ES = 0.15, trivial) increase in times for 10- and 30-meter sprints, respectively, indicating a decrease in performance. However, Harris and colleagues (16) reported a −2.3% (ES = −1.4, moderate) change in 10-yard sprint times, indicating improved performance after mixed load training. Intuitively, one would have expected Tricoli to have reported more favorable results in sprint speed, as the training prescription used in this study involved a combination of heavy load squats and plyometric movements. The integration of plyometrics into the training program provided greater eccentric loading and greater velocity specificity than the 30-80% 1RM loads prescribed by Harris and colleagues (16), and accordingly a more favorable sprint training response may have been expected.


A key point apparent in reviewing the literature is that the strength and conditioning professional must be cautious in applying research findings regarding the prescription of various power training loads and schemes. One factor making the application of research findings problematic is the variation in total training volume used in the training interventions studied. For example, Wilson and colleagues (34) investigated all three training modes, high load training, moderate load and plyometric training, in a study which is widely cited in the literature. Examination of the training parameters prescribed for these subjects showed considerable variation in the total training volume performed in each loading scheme. Both the high and moderate load groups were prescribed 4 sets of 6-10 repetitions. However, one group trained at the load that was proposed to maximize peak power (Pmax, ∼30% 1RM) and the other at a 6-10RM load (∼65-75% of 1RM) resulting in one group performing significantly greater training volume (sets × repetitions × load). The third group performed DJ training, which made quantification of comparative total training volume almost impossible. Given this issue, it would seem that comparison of training results between training groups is somewhat tenuous as the amount of overload provided by each training program is different. Future research needs to quantify the effect of program design on the nature and volume of overload during power training in more detail, and equate total training volume in some manner when investigating training loads.

In an attempt to provide a comparison between the 4 training approaches examined in this review, the number of moderate and large effect sizes for each of the variables discussed for each training technique is compared in Table 6. It can be observed from the table that heavy load explosive training is the most effective strategy of those investigated if a shift in maximum strength (1RM) in the squat movement is the desired training outcome. However, in and of itself, it is not the most effective loading pattern and/or exercise for the development of performance in sports-specific tasks such as jumping and sprinting. Moderate load explosive squat/JS training seems to be as effective as heavy load training at developing force parameters and effective in developing muscular power. Although moderate load training was the most effective load investigated in developing jump performance, the literature is still inconclusive as to its efficacy in developing sprint performance because only one moderate effect size was evident for this task. Although results relating to sprint performance should be interpreted in the context that a myriad of factors, other than the production of lower limb force and power effect sprint performance. Light load (BW) training seems the least effective of all the loading schemes investigated reinforcing the fact that the magnitude of the resistance is an important stimulus to adaptation. Mixed load training appears a promising loading scheme for improving force capability and functional performance, the ideal mixture of loading an area for future research.

Table 6
Table 6:
Comparison of number of moderate and large effect sizes for different loads in reviewed papers


The study of power is a major area of interest in sport and exercise science. Not surprisingly, therefore, the development of power has been the subject of a great deal of research and subsequent conjecture. Much of the conjecture can be attributed to the (a) great variation in methodologies among research; (b) lack of consistency between laboratories in terms of the rationale and execution of power assessment (1); (c) difficulty in identifying those training methods that best facilitate improvements in power; and (d) scarcity of research investigating the best methods of transferring power gains to sports specific tasks.

Reviewing the research into the assessment and development of power reveals a great deal of variation in the methodologies used by various researchers. The scope of this variation makes comparisons difficult and hence definitive conclusions practically impossible. For example, the vast majority of research has been relatively short in duration (8-12 weeks) and therefore the application of findings to long-term training is questionable as the influence of neural and morphological mechanisms change with training duration. Research in this area is also typified by a wide spectrum of loading parameters that include differences in (a) volume, (b) intensity-contraction force, (c) total work output, (d) tempo of concentric-eccentric contractions, (e) frequency, (f) rest/recovery time-density, and (g) type of contractions. Suffice to say, the strength and conditioning practitioner in selecting training loads must review the research, critically evaluating the aforementioned methodological inconsistencies. This evaluation must then be combined with practical experience and individual strength profiling of their athletes to apply appropriate load selection to their program design.

Nonetheless, cognizant of these limitations, the authors have tried to make sense as to which training loads best facilitate improvements in strength, power, and sports-specific performance through the use of effect statistics. As a result of this analysis, some broad conclusions are possible. It seems that heavy load training might illicit an improvement in the ability to generate high forces with some transference to power and little transference to functional performance such as jumping and sprinting. The use of moderate loading schemes appears the optimal load to maximize power and may contribute to gains in sports specific performance. Moderate load training appears particularly effective if ballistic techniques are used, i.e., JS. A mixed method approach (combination training), which is an integration of heavy and moderate, or heavy and light load training, appears a promising approach for developing the force and sports specific capability of muscle. There seems little benefit to use light weight, plyometric training in isolation. The findings of this review may prompt new insights into training practice and research directions. However, it more likely confirms the value of some of the practices already used by strength and conditioning coaches, whereby a variety of loads are utilized in a periodized approach to training based on training age, needs analysis and strength profiling of the athlete and competition structure.


1. Abernethy P, Wilson G, and Logan P. Strength and power assessment: Issues, controversies and challenges. Sports Med 19: 401-417, 1995.
2. Avela J and Komi P. Timing strategy of the eccentric stretching velocity of the leg extensor muscles in the drop jump exercises. J Biomechanics 27: 721-748, 1994.
3. Baker D, Nance S, and Moore M. The load that maximizes the average mechanical power output jump squats in power trained athletes. J Strength Cond Res 15: 92-97, 2001.
4. Bobbert MF, Mackay M, Schinkelshoek D, Huijing PA, and Schenau GJVI. Biomechanical analysis of drop and countermovement jumps. Eur J Appl Physiol 54: 566-573, 1986.
5. Chu D. Jumping into Plyometrics. Champaign, IL: Human Kinetics; 1998. pp. 1-5.
6. Cormie P, Deane R, and McBride JM. Methodological concerns for determining power output in the jump squat. J Strength Cond Res 21: 424-430, 2007.
7. Cormie P, McBride JM, and McCaulley GO. Validation of power measurement techniques in dynamic lower body resistance training. J Appl Biomech 23: 103-118, 2007.
8. Cormie P, McCaulley GO, Travis-Triplett N, and Mcbride JM. Optimal loading for maximal power output during lower-body resistance exercise. Med Sci Sports Exerc 39: 340-349, 2007.
9. Cronin J, McNair PJ, and Marshall RN. Force-velocity analysis of strength-training techniques and load: implications for training strategy and research. J Strength Cond Res 17: 148-155, 2003.
10. Delecluse C. Influence of strength training on sprint running performance. Current finding and implications for training. Sports Med 24: 147-156, 1997.
11. Delecluse C, Van Coppenolle H, Willems E, Van Leemputte M, Diels R, and Goris M. Influence of high resistance and high-velocity training on sprint performance. Med Sci Sports Exerc 27: 1203-1209, 1995.
12. Dugan EL, Doyle T, Humphries B, Hasson CJ, and Newton RU. Determining the optimal load for jump squats: a review of methods and calculations. J Strength Cond Res 18: 3, 2004.
13. Fatouros IG, Jamurtas AZ, Leontsini JD, Taxildaris K, Aggelousis N, Kostopoulos N, and Buckenmeyer P. Evaluation of plyometric exercise training, weight training, and their combination on vertical jumping performance and leg strength. J Strength Cond Res 14: 470-476, 2000.
14. Fatouros JG, Jamurtas A, and Buckenmeyer P. Plyometrics training and weight training effects on force-power parameters of vertical jumping. Med Sci Sports Exerc S174, 1983.
15. Gehri D, Ricard M, Kleiner D, and Kirkendall D. A comparison of plyometric training techniques for improving vertical jump ability and energy production. J Strength Cond Res 12: 85-89, 1998.
16. Harris GR, Stone MH, O'Bryant HS, and Proulx CM, Johnson RL. Short-term performance effects of high power, high force, or combined weight-training methods. J Strength Cond Res 14: 14-20, 2000.
17. Holcomb WR, Lander JE, Rutland MR, and Wilson GD. The effectiveness of a modified plyometric program on power and the vertical jump. J Strength Cond Res 10: 89-92, 1996.
18. Jones K, Bishop P, Hunter G, and Fleisig G. The effects of varying resistance-training loads on intermediate- and high-velocity-specific adaptations. J Strength Cond Res 15: 349-356, 2001.
19. Kaneko M, Fuchimoto T, Toji H, and Suei K. Training effect of different loads on the force-velocity relationship and mechanical power output in human muscle. Scand J Sports Sci 5: 50-55, 1983.
20. Kyrolainen H, Avela J, Mc Bride JM, Koskinen S, Andersen JL, Sipila S, Takala TES, and Komi PV. Effects of power training on muscle structure and neuromuscular performance. Scand J Sports Sci 15: 58-64, 2005.
21. Luebbers PE, Potteiger JA, Hulver MW, Thyfault JP, Carper MJ, and Lockwood RH. Effects of plyometric training and recovery on vertical jump performance and anaerobic power. J Strength Cond Res 17: 704-709, 2003.
22. Lyttle AD, Wilson GJ, and Ostrowski KJ. Enhancing performance: maximal power versus combined weights and plyometrics training. J Strength Cond Res 10: 173-179, 1996.
23. McBride JM, Triplett-McBride T, Davie A, and 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.
24. Moss B, Refsnes PF, Abildgaard A, Nicolaysen K, and Jensen J. Effects of maximal effort strength training with different loads on dynamic strength, cross-sectional area, load-power and load-velocity relationships. Eur J Appl Physiol 75: 193-199, 1997.
25. Murphy AJ and Wilson GW. The ability of tests of muscular function to reflect training-induced changes in performance. J Sports Sci 15: 191-200, 1997.
26. Neils CM, Udermann BE, Brice GA, Winchester JB, and McGuingan MR. Influence of contraction velocity in untrained individuals over the initial early phase of resistance training. J Strength Cond Res 19: 883-887, 2005.
27. Newton RU, Hakkinen K, Hakkinen A, McCormick M, Volek JS, and Kraemer WJ. Mixed-methods resistance training increases power and strength of young and older men. Med Sci Exerc Sports 34: 1367-1375, 2002.
28. Newton RU, Kraemer WJ, and Hakkinen K. Effects of ballistic training on preseason preparation of elite volleyball players. Med Sci Sports Exerc 31: 323-330, 1999.
29. Newton RU, Kraemer WJ, Hakkinen K, Humphries BJ, and Murphy AJ. Kinematics, kinetics and muscle activation during explosive upper body movements. J Appl Biomech 12: 31-43, 1996.
30. Potteiger JA, Lockwood RH, Haub MD, Dolezal BA, Almuzaini KS, Schroeder JM, and Zebas CJ. Muscle power and fiber characteristics following 8 weeks if plyometric training. J Strength Cond Res 13: 275-279, 1999
31. Rhea MR. Determining the magnitude of treatment effects in strength training research through the use of the effect size. J Strength Cond Res: 918-920, 2004.
32. Schmidtbleicher D and Haralambie G. Changes in contractile properties of muscle after strength training in man. Eur J Appl Physiol Occup Physiol 46: 221-228, 1981.
33. Tricoli VA, Lamas L, Carnevale R, and Ugrinowitsch C. Short-term effects on lower-body functional power development: weightlifting vs. vertical jump training programs. J Strength Cond Res 19: 433-437, 2005.
34. Wilson GJ, Newton RU, Murphy AJ, and Humphries BJ. The optimal training load for the development of dynamic athletic performance. Med Sci Sports Exerc 25: 1279-1286, 1993.
35. Young BW and Bilby EG. The effect of voluntary effort to influence speed of contraction on strength, muscular power, and hypertrophy development. J Strength Cond Res 7: 172-178, 1993.
No Caption Available

squat; jump squat; training load; power

© 2009 National Strength and Conditioning Association