Advances in powerlifting training methodology and gear (i.e., squat suit, bench press shirt, deadlift suit, knee wraps) have led to dramatic increases in current world records for the 3 lifts and a higher overall average of weights lifted at national and world competitions. Powerlifters typically train an average of 8–12 weeks per competitive cycle, with 3–4 workouts per week lasting anywhere from 1.5 to 3 hours per workout. It is not uncommon for athletes to wear modern suits and bench press shirts that are very tight and restrictive during these workouts.
Interestingly, recent research has shown that restricting blood flow while performing resistance training results in greater strength gains and muscle mass compared with nonoccluded training (4,10,11,14). This type of training is also referred to as KAATSU and is characterized by performing low-intensity cardiovascular or resistance training while occluding blood flow to the working muscles with bands or blood pressure cuffs (12). Although the exact mechanisms of these adaptations are unknown, one theory is that an accumulation of metabolites within the occluded muscle may stimulate subsequent increases in anabolic growth factors such as growth hormone, IGF-1, and testosterone (3,4). There is currently no research on the effects of blood flow restriction training on strength gains in recreational or competitive strength athletes engaged in regular, high-intensity training. This question is particularly important to competitive powerlifters who could apply this knowledge to their training by better calculating the optimal training volume performed while using competition equipment to see the largest performance gains.
The purpose of this study is to compare differences in performance gains between powerlifters who are both training and competing in equipment, thought to restrict blood flow, with those training and competing without such gear. It is hypothesized that lifters who train and compete in the restrictive equipment will be able to train with higher volumes than those training without equipment. Also, it is hypothesized that the equipped lifters will have greater increases in powerlifting performance on the squat, bench press, deadlift, and their total.
Experimental Approach to the Problem
To examine the effects of training with or without compressive powerlifting gear on performance, a randomized prospective design assessing training volume and performance outcomes between the 2 groups of lifters was performed. The intervention consisted of a 10-week strength training program. Dependent variables included training volume and performance measures for the Squat, Bench Press and Deadlift. Additional variables of interest included body composition and hemodynamic measures including arterial stiffness.
The main inclusion criteria for the participants were competitive collegiate powerlifters from the local community. Before testing, the subjects signed an informed consent approved by the Louisiana State University, Agricultural and Mechanical College Institutional Review Board. The participants varied in gender, body weight, and experience level. Upon agreeing to take part in the study, the lifters were randomly placed into either a group that trained and competed using compressive gear (CG) or without the gear (NON).
All physiological testing was performed in the morning hours (between 8 and 11:00 AM) in a climate controlled (23° C) exercise testing laboratory of the host institution. The subjects were instructed to fast and refrain from caffeine, tobacco, and strenuous activity for 12 hours before testing. Blood pressure, heart rate, and vascular stiffness were obtained from the participants before and after the training intervention. The subjects were instructed to rest quietly in a supine position for 20 minutes before hemodynamic assessments. Radial artery applanation tonometry and pulse wave analysis were used to calculate derived central blood pressures and central stiffness parameters (AtCor Medical). To capture pulse wave velocity, sequential measurements of arterial pressure waves at the carotid and posterior tibial arteries were made. The surface distances from the suprasternal (SS) notch to the carotid and peripheral sites were measured. Pressure wave transit times to each site were measured using the foot-of-the-wave method: PWV = distance (SS notch to peripheral site – SS notch to carotid site) / time (electrocardiogram [EKG] R-wave to peripheral pulse – EKG R-wave to carotid pulse. To assess the effects of the CG on limb blood flow, Doppler ultrasonography was performed on the popliteal artery of select participants in response to localized increases in occlusion pressure using a cuff, and using a squat suit and knee wraps, respectively. The varied cuff pressures were used to simulate differences in tight vs. loose fitting equipment. Body composition was assessed using skinfold calipers and the Jackson Pollock 3-site skinfold test to obtain body density (8,9). To estimate body fat, from the body density, the Siri equation was used (13). The 3 sites used for men were the chest, abdominal, and thigh. For women, the sites included the triceps, suprailiac, and thigh.
After preliminary hemodynamic testing, a simulated meet was conducted before the beginning of the intervention to gather 1 repetition maximum (1RMs) in the Squat, Bench Press, and Deadlift. As in a competition, lifting began at a set time and all attempts were judged according to standards set forth in the US Powerlifting federation rulebook (2). Also, the lifters performed the lifts in the same order as in a competition; beginning with 3 attempts in the Squat, then 3 attempts in the Bench Press, and ending with 3 attempts in the Deadlift. The highest successful attempt at each event was added together to get the total, which was recorded as the athletes performance premeasure and the highest successful individual attempts at each lift. At the end of the 10-week training intervention, all the lifters participated in a local powerlifting competition. The individual lifts that were passed as good lifts in competition, and the athletes total, were counted as performance postmeasures for the study. In addition, hemodynamic measures were assessed within 72 hours after the completion of the competition.
The Wilks coefficient is a formula developed by Robert Wilks of Australia that is used at powerlifting competitions to compare lifters of different body weights. Each lifter has a coefficient based on their body weight in either kilograms or pounds, which is then multiplied by his or her total to get a figure that is compared with other lifters in the meet to establish the best lifter in the competition. In competitions held in the United States, this figure is usually then divided by 2.2046 to get the lifters final score (1).
The intervention consisted of a 10-week powerlifting style training cycle that was developed to improve performance in the Squat, Bench Press, and Deadlift. The lifts of Squat and Bench Press were performed twice per week, with heavy Squats on Sunday along with a light Bench Press. Deadlifts were done once per week on Tuesday, and heavy Bench Press with a light Squat was performed on Thursday. The intensity of training (percentage of 1RM) was increased in a staggered progression, and started with 80% of the maximum for all competition exercises during the first 4 weeks of training. At week 5, the intensity was increased to 85% for 2 weeks before being increased to 90–95% during weeks 7 and 8. The lifters began to taper at week 9 by dropping the intensity to 85%, then down to 80% of the 1RM the first day of the tenth week. During the rest of week 10, training intensity was dropped to 70% to allow the lifter to taper for the competition. The training intensity for the cycle is shown in Figure 1.
Although the intensity followed a staggered linear progression model, the training volume (sets × reps × weight) followed a wave pattern of progression. During the first week, training volume began as 5 sets of 2 for heavy Squats and Bench Press, and 5 sets of 1 on the deadlift. During the next 3 weeks, volume was increased by adding 1 rep to each set per week, while maintaining the set number. Week 4 was the apex of volume, with 5 sets of 5 on the Squat and Bench Press, and 5 sets of 4 on the deadlift. At week 5, the volume was lowered by decreasing the set number to 4 and rep number to 2, and increased the next week to 4 sets of 3. At week 7, the volume was again lowered, this time by dropping the sets to 3, and reps to 1. At week 9, 3 sets of 2 reps were performed, which coincided with the highest intensity weeks of the training cycle. On Sunday of week 10, the volume was increased again to accommodate the tapering of intensity on that day only, and lowered along with the intensity the rest of the week to allow for the lifter to recover for the competition.
The program followed the physiological principle of specificity, with the majority of the training consisting of the competition lifts. The equipped group performed all work sets of 80% or greater wearing competition gear, while the unequipped group was only allowed to wear a weight belt on work sets. Even though the equipped group performed all of their work sets with equipment, they performed auxiliary work such as the light Bench Press on Sunday and the light Squat on Bench Press day sans equipment. Also, other auxiliary exercises were performed unequipped on each training day in addition to the competition lifts that were meant to compliment the heavy lift of that day. For example, heavy Squat was performed on Sunday, followed by light Bench Press, then a different squatting exercise to build additional strength in the legs. On Tuesday, the lifters competition style Deadlift was performed as the main exercise, followed by a different style Deadlift as a compliment to the main lift. On Thursday, after heavy Bench Press and light Squat, a second type of Bench Press was performed to further strengthen the upper body pressing muscles. These auxiliary exercises were performed at a much lower intensity and slightly higher reps than the competition lift, and done without equipment in both groups.
Statistical analyses were performed using SPSS (18.0). Group values are expressed as mean ± SD. Pretraining group differences were examined using independent t-tests. Differences in training volume across the 10 weeks were examined using a 2 (CG and NON groups) × 10 (weeks of training) analysis of variance with repeated measures. The influence of training on performance measures was examined using a multivariate analysis of variance. An Alpha of 0.05 was required for statistical significance.
Eighteen adults (14 men and 4 women) between the ages of 18 and 26 years participated in this study. The CG group consisted of 6 men and 2 women, and the unequipped (NON) group consisted of 8 men and 2 women. Participant characteristics are presented in Table 1. On average, the lifters in the CG group were heavier than those in the NON group (CG: 198.34 ± 49.64 lb vs. Non: 170.9 ± 42.56 lb, p = 0.23), but the difference did not reach statistical significance. The NON group had a slightly lower body fat percentage than the CG (CG: 19.36 ± 11.38% vs. non: 14.02 ± 4.66%) With body composition taken into account, the NON lifters had a similar lean body mass (LBM) as the CG (CG: 150.42 ± 22.88 vs. Non: 158.6 ± 33.09 lb), but averaged one weight class lower (CG: 195 ± 43.44 vs. Non: 171 ± 44.76 lb).
No differences were detected between the groups in terms of hemodynamic variables (CG: [SBP] 121 ± 12.37 and [DBP] 69 ± 8.08 vs. Non: [SBP] 117 ± 12.47 and [DBP] 67 ± 5.59 mm Hg, p > 0.05). No group differences in heart rate (CG: 77 ± 7.37 vs. Non: 72 ± 13.89 b·min−1, p > 0.05), PWV (CG: 7.07 ± 0.49 vs. Non = 7.48 ± 1.17 cm·s−1, p > 0.05), or the augmentation index (CG: 3.25 ± 8.10 vs. Non: 3.90 ± 8.85, p > 0.05) were detected.
Figures 2A–D depict the progression of volume lifted during the 10-week intervention, for the squat (Figure 2A), bench press (Figure 2B), the deadlift (Figure 2C), and the combined volume for the 3 lifts (Figure 2D). There was a significant increase in the volume of weight lifted for each exercise during the base preparation phase for both groups. During this phase, the volume lifted for the squat and the totals showed a greater volume increase in the CG group. However, after the base preparation phase, the progression of volume lifted for all the exercises was quite similar between groups, for the remainder of the training cycle.
Table 2 presents the performance measures for the 3 required lifts: squat, bench press, deadlift, and the totals score before and after the 10-week intervention. There was a significant increase in squat (↑19.05 ± 30.97 lb, p = 0.02), deadlift (↑19.05 ± 21.17 lb, p = 0.001), and the total score (↑44.00 ± 60.44 lb, p = 0.005) for both groups combined. Those in the CG group had greater improvements in the squat lift (CG: 33.85 vs. Non: 5.74 lb, p = 0.07), and the Totals score (CG: 66.59 vs. Non: 23.67 lb, p = 0.15). However, the difference did not reach the a priori alpha level of p < 0.05. No trends were noted between the groups for the other performance measures, even when adjusting for preintervention strength differences, or body weight.
Both groups combined lost an average of 2.67 lb of body weight after the intervention (p = 0.06). No group differences were observed. Both groups combined showed a significant increase in the Wilks scores (+13.54 points, p = 0.03). However, there were no significant differences between the groups.
The findings of the study indicate a significant increase in powerlifting performance after the 10-week training cycle. Uniquely, this study also indicates those individuals who trained using CG had a greater improvement in the squat and the combined total of the three lifts. Finally, both groups showed a significant improvement in their Wilks scores, indicating the efficacy of the training program.
The participants in this study were collegiate powerlifters recruited from the Louisiana State University powerlifting club with experience levels ranging from 1 to 10 years. Only 1 participant had 10 years of lifting experience, whereas the majority had been directly involved in the sport for 1–4 years. There was 1 male collegiate national champion (former or current) in the CG group and 1 female collegiate national champion in the NON group. Besides those 2 top ranked lifters, the NON group contained 3 collegiate All American lifters (Nationally ranked in the top 3 of their weight class) compared with 1 All American in the CG group.
The body composition of the athletes in this study is fairly typical for this population. On average, the CG lifters were heavier than the NON lifters by 27.44 lb, although the NON lifters had less body fat than did those in the CG group (Non: 14% and CG: 17.64%, respectively). When taking body composition into account, LBM between the groups was virtually identical; however, the CG group averaged one weight class higher. This indicates that the average lifter in the NON group had slightly greater LBM than the CG lifters.
The data in this study revealed several interesting trends regarding the cardiovascular health of competitive powerlifters. The average PWV for this group of lifters is considered normal for this age group. In fact, several participants measured <6 m·s−1, a value indicative of high arterial compliance, and excellent vascular health (AtCor Medical). Others report an inverse relationship between bench press strength and aortic stiffness, with the strongest in the group having the lowest central and peripheral PWV (7). It is important to understand there is no consensus in the literature regarding resistance training and vascular health (6,7). Resting heart rates for the study participants were also in the normal range for their age group, although generally higher than those observed in endurance athletes of the same experience level. Combined, the hemodynamic data in this study do not indicate there are deleterious vascular effects associated with this sort of training. We do recognize the need for further studies to examine the effect of chronic high-intensity strength training on cardiovascular biology.
The purpose of the first 4-week mesocycle was to build a foundation of strength by using training loads of 80% of the lifters 1RM throughout, and gradually increasing reps per set each week until the apex of volume was reached during the fourth week. According to a recent meta-analysis (16) aimed at examining the optimal training intensity for maximal strength gain, athletes with ≥1 year of experience exhibit the most gains with workloads of 80% of their 1RM (16). Many programs developed by successful strength athletes and coaches reflect this idea, such as those whose programs influenced the one in this study, namely, Dr. Fred Hatfield (first powerlifter to squat over 1,000 lb in competition) and highly successful Russian Powerlifting Coach Boris Sheiko. However, although the intensity during this training phase was calculated at 80% of the lifters 1RM, a subjective rating of perceived exertion scale was used by the coaches to determine if the weight was a correct reflection of this intensity. For example, if the lifter reported a rating of 1 (Easy) after the first work set, a supervising coach would adjust the bar weight by approximately 2–3% for the subsequent sets. If the lifter reported a 3 (Hard) the weight would be decreased on subsequent sets by 2–3%.
The results of the study indicate that there were no statistical significant group differences regarding the rate of volume increase for any of the lifts or the totals. Perhaps interestingly, it appears that the rate of change in volume lifted for the squat was greater in the CG group (CG: 5,062.5 vs. Non: 4,396.36 lb). Recognizing that there were no significant differences between the groups for the squat, we do appreciate that small changes in athletic training may translate to small yet important changes in performance. For example, examination of the percent difference in performance between first and fourth place in elite track and field athletes, found variances in performance as low as 1% enough to explain the difference between a gold medal and fourth place (5). Typically, such a small difference will not result in statistical significance and the fact that there was no statistical difference makes it hard to suggest any explanation as to why the CG group had a slightly greater magnitude of change in volume lifted.
After the apex of volume was reached during week 4, the focus of the program shifted from establishing a foundation to preparing for competition. For this reason, the total volume was decreased by lowering the repetitions and sets from 5 sets of 5 reps @ 80% to 4 sets of 2 reps @ 85% during week 5. This decrease in volume allowed for a period of delayed transformation in which the lifter could recover from the volume accumulation during the base preparation phase, while simultaneously adapting to heavier loads (85–95% of 1RM) that would prepare them for competition. As the competition approached, specificity was emphasized by increasing the intensity of the main lifts to 90–95%, while decreasing the total volume (sets and repetitions) to 3 sets of 1–2 reps. Because of the loads being close to the lifters' 1RM during this period, special care was taken by the coaches to make sure that the lifters stayed below 100% intensity. Therefore, the participants were rarely allowed to increase the bar weight above the prescribed intensity for that session, as was allowed during the base preparation phase. Weeks 9 and 10 was the tapering period in which both volume and intensity were decreased, and it is hypothesized that during this period of delayed transformation, a rebound effect is achieved when the lifter heals from the previous weeks of intense training (15).
The performance improvements, in this study, are indeed noteworthy, considering the program was only 10 weeks. In fact, the average improvements of the participants in this study were similar to those reported for nationally ranked collegiate powerlifters, over a period of 1 year. Specifically, the annual performance improvements of 5 All American (AA) U.S.A.P.L collegiate powerlifters, tracked across their 4-year Collegiate National powerlifting careers, averaged 32.45 lb in the squat, 18 lb in the Bench Press, and 12.58 lb in the Deadlift (2). The combined group improvement for squat in this study was only 13.4 lb less than the average AA, and the deadlift gain actually surpassed the annual improvements of the AA lifters after only 10 weeks of training. Furthermore, the performance improvements of the CG lifters in this study almost matched the annual squat gains of the AA and nearly doubled their gains in the deadlift. Although statistical differences were not detected between the groups in the performance measures, the group difference in the squat (CG: 33.85 vs. Non: 5.74 lb, p = 0.07) approached statistical significance. Improvements in totals were also greater in the CG lifters (CG: 66.59 vs. Non = 23.67 lb, p = 0.15) and also did not reach statistical significance at the a priori set alpha.
We believe the above difference is perhaps worthy of additional discussion. A possible explanation for the greater performance gain in squat performance in the CG group may be, in part, because of blood flow restriction to the legs while squatting in the CG. Most studies that have examined the effects of blood flow occlusion training have found higher strength gains (4), albeit with possible negative vascular consequences (6). To examine whether the suit and knee wraps altered blood flow kinetics in the powerlifters, blood flow velocity (FVI) was measured in the popliteal artery during 4 conditions: (a) Supine without suit and wraps, (b) Supine with Suit, and (c) Supine with Suit and knee wraps. At baseline, FVI averaged 3.43 cm·s−1 through the popliteal artery. With the addition of a competition squat suit, FVI dropped to 1.84 cm·s−1, suggesting a significant reduction in popliteal flow. With the addition of competition knee wraps the FVI decreased to 0 cm·s−1, suggesting full occlusion.
Recognizing the above measurements were obtained at rest, we suspect that during a squat performance that results in a significant pressor response there may be some blood flow to the area but argue there would still be significant restriction. The fact, the greatest difference in strength gains were seen with the squat performance would support previous research which indicates that the effects of blood flow restriction training is specific to the muscles involved in the movement. In contrast, the Bench Press shirt, although quite restrictive, may not be sufficient to occlude blood flow to prime movers of the pectorals, triceps, and anterior deltoids during the Bench Press lift. In fact, the purpose of the Bench Press shirt is to assist the lifter during the eccentric phase of the lift, serving to give the lifter a “rebound” effect during the concentric phase. We suspect this action may not result in significant blood flow restriction to the working muscles. Coincidentally, similar gains in performance were seen between equipped and nonequipped lifters in the Bench Press at post testing.
Regarding the Deadlift, many lifters experience little to no benefit when using CG for this lift; therefore, the lack of difference between the groups is not surprising. The Deadlift is an eccentric-less lift that begins from the floor and cannot use the exaggerated stretch reflex provided by the equipment of the Squat and Bench Press. Additionally, success in the Deadlift is largely influenced by the correct body position that allows for the greatest possible leverage. Specifically, the lifter must be able to achieve a neutral spine with shoulders back and down at the start of the lift, which is made difficult with the use of an extremely tight suit. Furthermore, because most lifters use either a Squat suit for Deadlifting, or a suit almost identical in form to the Squat suit, it can be assumed that not >50% blood flow occlusion to the popliteal artery alone takes place during the lift, which, similar to the Bench Press shirt, does not qualify an equipped Deadlift as an occlusion exercise that can be compared with methods used in other blood flow restriction studies.
The significant increase in the Wilks score is another indicator of the success of the program. The Wilks score is used in powerlifting competitions to compare the strength of lifters in different weight classes; it is used to award the best lifter of the meet. Because the Wilks score takes body weight into account, it is possible for increases in strength to occur even if the absolute weight lifted does not change.
A potential limitation of this study includes the heterogeneous makeup of the population with regards to competitive experience (1–10 years.) and skill level (novice to elite) makes it difficult to contribute group differences in performance solely to the equipment. Coincidently, there was a slight group difference in skill level, with the CG group containing slightly more novice lifters while the NON group contained more intermediate. Very few elite lifters participated in the study, who may have been less affected by the technical aspects of the equipment, precompetition anxiety, and other variables that affect performance than their novice counterparts. Another potential limiting factor in this study may have existed in the training design itself. Specifically, both groups performed auxiliary exercises in addition to the main lift of the day that was meant to improve on the weaknesses and muscular imbalances in the lifter. These auxiliary exercises included variations of the main lifts, which were always done without equipment in both groups. In fact, approximately 33% of a workout was performed in the CG. Although a trend has emerged suggesting that the CG group was handling a greater amount of volume in the squat, this may have been more pronounced if more total volume in the equipped group had been performed in the equipment.
The ergogenic aid of the powerlifting gear may be important for other athletic populations, recreational trainers, and clinical populations alike. The fact that the biggest performance increase in this study involved the CG squat lift is a topic worthy of further investigation by the strength and conditioning community. Additionally, it is a common point of argument among powerlifters and their coaches as to the amount of training that should be performed in competition equipment to see the greatest gains. Because this is the first study to compare CG and nonequipped training, it may be used as a starting point for further studies to examine this topic in greater detail and contribute to current training knowledge on the art of strength training. Additionally, this study adds to the literature on the benefits and possible limitations on blood flow occlusion training, and its application to strength athletes or even clinical populations. Almost all of the current research that examines occlusion training uses low intensities in their protocols, and none have used experienced powerlifters who regularly lift at very high intensities (80–100%) while the muscles in the working limbs are in an occluded state.
The authors would like to thank the Louisiana State University Powerlifting Club and the graduate students of the Department of Kinesiology for their participation and technical support. None of the authors have a conflict of interest (financial or otherwise) in the subject matter of this article. This Study was approved by the Institutional Review Board of the Louisiana State University, Agricultural and Mechanical College.
1. International Powerlifting Federation: IPF, June 2011. Available at: http://www.powerlifting-ipf.com/
. Accessed May 5, 2011.
2. U.S.A. Powerlifting—The Choice for Drug Free Strength
Sport. 2011. Available at: http://www.usapowerlifting.com/index.shtml
. Accessed May 5, 2011.
3. Abe T, Kearns CF, Sato Y. Muscle size and strength
are increased following walk training with restricted venous blood flow from the leg muscle, Kaatsu-walk training. J Appl Physiol 100: 1460–1466, 2006.
4. Abe T, Sakamaki M, Fujita S, Ozaki H, Sugaya M, Sato Y, Nakajima T. Effects of low-intensity walk training with restricted leg blood flow on muscle strength
and aerobic capacity in older adults. J Geriatr Phys Ther 33: 34–40, 2010.
5. Andre MJ, Winchester JB, Hartman MJ, Nelson AG. Identifying meaningful differences in elite athletes. Med Sci Sports Exerc 42: 787, 2010.
6. Credeur DP, Hollis BC, Welsch MA. Effects of handgrip training with venous restriction on brachial artery vasodilation. Med Sci Sports Exerc 2009.
7. Fahs CA, Heffernan KS, Ranadive S, Jae SY, Fernhall B. Muscular strength
is inversely associated with aortic stiffness in young men. Med Sci Sports Exerc 42: 1619–1624, 2010.
8. Heffernan KS, Fahs CA, Iwamoto GA, Jae SY, Wilund KR, Woods JA, Fernhall B. Resistance exercise training reduces central blood pressure and improves microvascular function in African American and white men. Atherosclerosis 207: 220–226, 2009.
9. Jackson AS, Pollock ML. Generalized equations for predicting body density of men. Br J Nutr 40: 497–504, 1978.
10. Jackson AS, Pollock ML, Gettman LR. Intertester reliability of selected skinfold and circumference measurements and percent fat estimates. Res Q 49: 546–551, 1978.
11. Loenneke JP, Wilson GJ, Wilson JM. A mechanistic approach to blood flow occlusion. Int J Sports Med 31: 1–4, 2010.
12. Meyer RA. Does blood flow restriction enhance hypertrophic signaling in skeletal muscle? J Appl Physiol 100: 1443–1444, 2006.
13. Peterson MD, Rhea MR, Alvar BA. Maximizing strength
development in athletes: a meta-analysis to determine the dose-response relationship. J Strength
Cond Res 18: 377–382, 2004.
14. Sato Y. The history and future of KAATSU training. Int J Kaatsu Train Res 1: 1–5, 2005.
15. Siri WE. Body composition from fluid spaces and density: Analysis of methods. 1961. Nutrition 9: 480–491, 1993; discussion 480, 492.
16. Yasuda T, Ogasawara R, Sakamaki M, Ozaki H, Sato Y, Abe T. Combined effects of low-intensity blood flow restriction training and high-intensity resistance training on muscle strength
and size. Eur J Appl Physiol 111: 2525–2533, 2011.
17. Zatsiorsky VM, Kraemer WJ, ed. Science and Practice of Strength
Training (2 ed.). Champaign, IL: Human Kinetics, 2006.