Journal of Strength and Conditioning Research, 2000, 14(3), 241±247 q 2000 National Strength & Conditioning Association Comparative Effects of Deep Versus Shallow Squat and Leg-Press Training on Vertical Jumping Ability and Related Factors LAWRENCE W. WEISS,1 ANDREW C. FRY,1 LARRY E. WOOD,1 GEORGE E. RELYEA,2 AND CHARLIE MELTON1 1Musculoskeletal Dynamics Laboratory, Human Performance Laboratories, Department of Human Movement Sciences & Education, The University of Memphis, Memphis, Tennessee 38152-3480; 2Statistical Services, The University of Memphis, Memphis, Tennessee, 38152-3480. ABSTRACT Introduction Young, previously untrained healthy men (n 5 10) and wom- ertical jumping ability plays an important role in 5 en (n 8) completed 9 weeks of periodized, machine-based the level of success attained by individuals partic- squat training to determine if manipulating range of motion V would have a differential effect on vertical jumping ability and ipating in many sports and recreational activities. Each related measures. Subjects were pretested and then randomly person's ability to jump depends on a combination of assigned to 1 of 3 groups: (a) deep squats (n 5 6), (b) shallow other physical attributes (e.g., power, strength, and squats (n 5 6), and (c) controls (n 5 6). Training took place body composition) as well as the nature of actions im- 3 days per week. Pre- and posttesting included standing (RVJ) mediately preceding jumping (e.g., counter or non± and depth (DVJ) vertical jumps for distance; machine deep countermovement, running or standing start, 1- or 2- and shallow squats for 1RM (1 repetition maximum) relative legged takeoff). Because the requirements for various 21 strength; and velocity-controlled squats at 0.51 m´s for rel- vertical jumps are different and each person's combi- ative peak force and at 1.43 m´s21 for relative peak power. nation of physical attributes undergirding jumping is Based on ANCOVA posttest results, the training protocols were ineffective in eliciting improved performance (p . 0.05) also different, the performance on 1 type of jump will in VJ, slow-velocity squatting force, and moderately fast squat- not necessarily re¯ect performance on all types of ver- ting power when performance was compared with the per- tical jumps. Weiss et al. (12, 13) identi®ed several fac- formance of control subjects. Conversely, the group training tors in young adults that account for much of the vari- with deep squats was the only group to perform signi®cantly ability in 2 types of vertical jumps: a depth or box (p , 0.05) better than controls for 1RM shallow squats and vertical jump (DVJ) and a restricted-motion standing signi®cantly (p , 0.05) better than both shallow-squat and vertical jump (RVJ). For DVJ (box height 5 20 cm), the control groups for 1RM deep squats. Furthermore, the coef- most important predictor was relative peak squatting ®cient of transfer for deep squats to both RVJ (2.32) and DVJ 21 (1.68) was substantially greater than for shallow squats (0.31 power (peak power per body weight) at 1.43 m´s . and 0.11, respectively). It was concluded that deep-squat train- Predicted DVJ was then attenuated using a constant ing appears to elicit the best improvement for both shallow- for all female subjects. No such adjustment was made and deep-squatting performance. However, 9 weeks of ma- for male subjects. For RVJ, the most important predic- chine-based, periodized squat training, regardless of depth, tor was also relative peak squatting power at 1.43 does not appear to appreciably enhance slow-velocity squat- m´s21, after which body fat percentage and relative ting force, moderately fast squatting power, or vertical jump- peak squatting force (peak force per body weight) at ing distance in previously untrained men and women. 0.51 m´s21 were used in that order to attenuate the pre- Key Words: weight training, force, power dicted RVJ distance. Individuals experiencing changes in 1 or more of these predictors would be expected to Reference Data: Weiss, L.W., A.C. Fry, L.E. Wood, G.E. simultaneously experience corresponding changes in Relyea, and C. Melton. Comparative effects of deep jumping performance and vice versa. versus shallow squat and leg-press training on vertical Based upon the widely held premise of speci®city jumping ability and related factors. J. Strength Cond. of training, exercises having movement patterns most Res. 14(3):241±247. 2000. similar to vertical jumping should elicit the best im- provements in jumping performance (2). The squat 241 242 Weiss, Fry, Wood, Relyea, and Melton and possibly leg press exercises incorporate movement deep dynamic constant external resistance (DCER) patterns at the ankle, knee, and hip that are similar to squats in an effort to stabilize the learning process that those used in vertical jumping. Furthermore, since the normally accompanies exposure to new motor skills countermovement used in vertical jumping typically (in this case, skills for which we assessed strength and occurs through a limited range of motion, it follows power) (9). that a shallow squat-training depth would likely elicit Subsequent to the 9-week training period and im- the best improvements in jumping performance. In mediately following the last training session, subjects practice, both deep and shallow squats are used for practiced each of the aforementioned tests of muscle this purpose, and have been categorized by Baker (2) function. This procedure was used to refamiliarize the as ``general'' strength training. In this regard, he de- subjects (as well as the control group) with the stan- ®ned general strength-training exercises as ``those dardized protocols they had not practiced for 9 weeks. aimed at increasing the maximal strength of the mus- Body Fat Measurement. Body fat estimation was per- cles involved in jumping. Examples would be squats, formed during pre- and posttesting periods. A 3-site, front squats, split squats, and power shrugs (with a gender-speci®c skinfold test (8) was exclusively per- very slow eccentric dip to the knee).'' It appears gen- formed by the primary investigator to estimate body eral strength training in and of itself may elicit mar- fat percentage. All sites were marked with water-sol- ginal or no improvements in jumping performance in uble ink and subsequently measured 3 times in series. elite athletes, even when strength is substantially im- Skyndex calipers (Caldwell, Justice and Co., Fayette- proved (1). It may, however, elicit signi®cant improve- ville, AR) were checked for a ``zero'' starting position ments in vertical jumping by lesser-trained individuals at every session in which skinfolds were obtained, al- (3, 7). though jaw-width calibration was not assessed. A Ster- Considering the aforementioned comments, the ling electronic platform scale (Sterling Scale Co., overall purpose of the current investigation was to de- South®eld, MI) was used to obtain body weight. The termine if squatting depth during short-term, period- device was calibrated before and then checked daily ized squat training would differentially impact jump- during testing using a certi®ed 50.00-kg weight. An ing performance and 1 or more predictors of VJ ability. interday coef®cient of variation of 0.0% was found for More speci®cally, the purpose of this investigation was calibration tests performed over a 10-day span. Sub- to determine the comparative effects of shallow versus jects were weighed in their testing attire (shorts, socks, deep squat training on vertical jumping performance shoes, and a T-shirt). (RVJ and DVJ), on variables reported to be signi®- Vertical Jumps. The depth or box jumps and the cantly related to jumping, and on other factors asso- standing or restricted vertical jumps were performed ciated with squatting strength. as previously described (12, 13). For the DVJ, the sub- Methods ject stepped off a 20-cm box and rebounded from the ¯oor to enhance the prestretch occurring during the Subjects eccentric or countermovement phase. (An analogous Healthy young male (n 5 10) and female (n 5 8) uni- prestretch occurs when an individual takes a running versity students, averaging 23.7 years of age (SD 5 start and ``blocks'' prior to jumping.) The RVJ also in- 6.1), weighing an average of 78.9 kg (SD 5 24.4), and corporated a countermovement but with substantially having a wide array of vertical jumping abilities vol- less intense prestretching. (A relatively smaller pre- untarily served as subjects in this investigation. Sub- stretch normally occurs when an individual chooses to jects had not engaged in any formal strength devel- jump without a running approach.) A Vertec device opment program for a minimum of 1 year prior to the (Power Systems Inc., Knoxville, TN) was used for mea- study. Subsequent to both written and verbal expla- suring vertical jumping distance to the nearest 0.5 inch nations of the speci®c nature of each volunteer's in- (subsequently converted to meters). volvement in the study, a medical history question- Velocity-Limited Squats. Concentric-only squats (pre- naire was completed and written informed consent ceded by a nonresisted countermovement) were per- was obtained as approved by the University's Institu- formed at slow (0.51 m´s21) and moderately fast (1.43 tional Review Board. Finally, all volunteers were m´s21) velocities as previously described (11) using an screened by auscultation for both hypo- and hyperten- Ariel 5000 multifunction dynamometer computerized sion prior to their designation as subjects. exercise system (CES) (Ariel Dynamics Inc., San Diego, Testing Protocols CA). Of primary interest were relative peak power Familiarization. Over a 2-week period and prior to pre- generated at 1.43 m´s21 and relative peak force gener- testing, 6 50-minute familiarization sessions took ated at 0.51 m´s21 because these variables were previ- place.