Secondary Logo

Journal Logo

Original Research

The Relation Between Anthropometric and Physiological variables and Bat Velocity of High-School Baseball Players Before and After 12 Weeks of Training

Szymanski, David J1; Szymanski, Jessica M1; Schade, Ryan L2; Bradford, T Jason2; McIntyre, Joseph S3; DeRenne, Coop4; Madsen, Nels H3

Author Information
Journal of Strength and Conditioning Research: November 2010 - Volume 24 - Issue 11 - p 2933-2943
doi: 10.1519/JSC.0b013e3181f0a76a
  • Free



Strength and power are integral components of a baseball player's offensive and defensive performances on the field. In an attempt to improve these 2 components, players are resistance training, throwing medicine balls, performing plyometrics, and swinging over and underweighted bats (8,9,15,32-34). Research suggests that the biomechanics of swinging a baseball bat with greater velocity has strong relations with fitness and performance tests (2,5,19,26,27,29,31,38). Baseball research suggests that bat swing velocity (BV) is an important variable affecting how hard the ball is hit, how far the ball travels, or both (1,6,18). Therefore, BV represents 1 baseball-specific skill variable that strength and conditioning coaches may strive to increase in baseball hitters to provide them the opportunity to enhance game performances. It must be stated that even though a hitter may increase his BV, this does mean that he will hit for a higher batting average or slugging percentage (power). Hitters' goals always include attempting to hit the pitched baseball on the center of percussion (sweet spot) safely with greater BV. Thus, identifying potential exercises that may increase BV and field tests for baseball players that relate to the baseball-specific skill of BV would be useful for those that train them.

It has been reported in biomechanical (11,12,40), cinematographic (18), and electromyographical (24) studies that hitting a baseball is a sequence of coordinated muscle activity connected together by the hips, torso, and arms (kinetic chain). These researchers suggest that those that train baseball players should have them focus on performing multijoint leg and explosive hip and torso rotational exercises. It has been reported that BV significantly increased after a period of resistance training for high-school baseball players (30,33,34), or after using under and overweighted implement training for high school (8) and college (9,23) baseball players.

In addition to the strength of the muscles activated during the baseball swing, the power of the active musculature of the legs and trunk is also likely to be an important factor affecting BV. Increased mechanical power will allow greater mechanical work to be transferred to the bat during the swing per unit of time, thus increasing BV. Previous research has reported that lower body power (27,29), upper body power (29), and torso rotational power (5) are correlated to BV of adolescent, high-school, and college baseball players. Furthermore, it has been reported that improvements in angular hip velocity (AHV) related to improvements in linear BV (33). Most recently, Hoffman et al. (14) indicated that lean body mass (LBM), speed, lower body power, and grip strength correlated to baseball-specific variables of home runs, total bases, slugging percentage, and stolen bases.

In strength and conditioning research, there is debate as to which is the most effective way to produce baseball-specific strength and power. Until more research is completed on this topic, this debate is a mute point because discussions may only be based on practical experience or anecdotal evidence. Although real-world experiences are valuable, scientific data that use valid and reliable equipment that measures BV consistently and accurately is needed to assist strength coaches in making important decisions on how to most effectively train their athletes in the weight room to improve this baseball-specific skill. For those who read research articles or attend conferences where this issue is discussed, 2 questions remain, “How should strength and conditioning coaches use or interpret correlational data on BV?” and “Do the correlations with BV mean a cause and effect relationship?”

The purpose of this article is to report on the relation between anthropometric (height, body mass, percent body fat [%BF], and LBM) and physiological (rotational power, rotational strength, vertical jump [VJ], estimated peak power, upper and lower body strength, and angular hip and shoulder velocities) variables to linear BV before and after completing a 12-week resistance training program of high-school baseball players. It is thought that this information will be of importance to strength and conditioning professionals in analyzing and planning the resistance training program for high-school baseball players with potential implications to college and professional baseball players. Furthermore, this article will also discuss how to interpret the correlation data by comparing it to reanalyzed 12-week pre and postresistance training data (33). However, this article will not discuss the effect of the 12-week resistance training program on BV because this has already been reported (33).


Experimental Approach to the Problem

A correlational study design was used to assess the magnitude of the relationships between various anthropometric and physiological performance variables to linear BV of high-school baseball players. Correlation analysis was chosen as the study design because we were interested in investigating the magnitude of the relationship between height, body mass, %BF, LBM, upper and lower body strength assessments, lower body power, and angular hip and shoulder velocities to linear BV and then comparing these relationships to 12-week resistance training results. We assigned the anthropometric and physiological performance measures as independent variables and linear BV as the dependent variable to establish anthropometric and physiological variable's relationship to linear BV. Linear BV was evaluated because it is a component of successful hitting (6,18) and the velocity of a hitter's swing at contact will make the ball travel farther, exit the bat with greater batted-ball velocity, or both (18).


Participants included 49 male baseball players from a 4A high-school state champion team from Columbus, GA, USA. A Descriptive Data Questionnaire indicated that subjects had been playing the sport of baseball for 9.5 ± 1.2 years and playing high-school baseball for 2.4 ± 0.6 years. All players were medically cleared by their family doctor before attending their first day of high school. A modified Physical Activity Readiness Questionnaire was also given to assess their health. If players progressed through both screenings, they were allowed to participate in the study. There were no declared health problems. Before participating in the testing session, which included being tested for 3 repetition maximum (RM) for parallel squat (PS) and bench press (BP), all players participated in 2 low-resistance (5× 5RM) familiarization strength training sessions to demonstrate proper lifting and spotting techniques to the investigators. All procedures were approved by a university institutional review board, and all subjects ≥18 years of age or subjects and parent or legal guardian provided written informed consent for subjects <8 years of age before participation.

Participants were separated by academic grades (freshman, sophomore, junior, and senior) and body mass categories (45.4-58.6, 59.1-72.3, 72.7-85.9, 86.4+ kg), which were modified from boys' wrestling weight classifications (36). To keep the number of participants per group large enough for statistical power and maintain homogeneity as a function of age and body mass, the participants were randomly assigned to 1 of 2 training groups by a stratified sampling technique. Group 1 (n = 24) and 2 (n = 25) both performed a stepwise periodized full-body resistance exercise program and took 100 swings a day, 3 d·wk−1, with their normal game bat for 12 weeks during the off-season (fall semester). Group 2 performed additional rotational and full-body medicine ball exercises 3 d·wk−1 for 12 weeks. The participants had to attend 90% (n = 33) of the total 36 exercise sessions to be included in the study. Participants could not miss exercise sessions on subsequent days or they would be dropped from the study. For control of outside influences, all participants were instructed to consume a normal diet and abstain from additional resistance training and taking ergogenic aids (e.g., creatine, amino acids, metabolite β-hydroxy β-methylbutyrate monohydrate) during the 12-week research period. Each participant recorded pre and posttesting food and drink consumption the day before and the day of testing in a Diet Log to assure that his normal diet was maintained.


Fourteen anthropometric, physiological, and baseball-specific skill variables were measured before and after completing a 12-week resistance training program. Descriptive data were collected for all subjects, including height, body mass, %BF, and LBM. Table 1 presents group 1 and 2 means and SD of anthropometric data for pre and posttraining. A 3-site skinfold test was used to determine %BF and LBM. Angular hip velocity and angular shoulder velocity (ASV) and baseball-specific (linear BV) data were collected for all subjects, which were measured by a motion capture system that identified and digitally processed reflective markers attached to participant's bat and body. Torso rotational strength was determined by measuring dominant and nondominant torso rotational strength (NDTRS) by a Cybex™ torso rotation machine. Sequential hip-torso-arm rotational strength was measured by a medicine ball hitter's throw (MBHT). Vertical jump was measured with a Vertec™ apparatus, and estimated peak power was calculated from the Sayers et al. (22) countermovement peak power equation. Parallel squat and BP were measured by a 3RM strength test. One repetition maximum for both tests was then calculated using a load assessment table by Wathen (39). All the tests were conducted on 1 day at the same time in an air-conditioned facility during the off-season when no team baseball practice was allowed because of state high school rules.

Table 1
Table 1:
Pre and posttraining anthropometric data for high-school baseball players.*†


Skinfold thickness was measured on the right side of the body with a calibrated Lange™ skinfold caliper (Beta Technology Incorporated, Cambridge, MA, USA) by an investigator who had previously demonstrated a test-retest reliability of r > 0.90 (p ≤ 0.05). Measurements were taken according to the recommendations of Jackson and Pollock (17) at the chest, abdomen, and thigh for the men. The investigator rotated through the measurement sites using a tape measure for proper location once and then repeated the measurements a second time in rotation. If the measurements were not within 1 mm, the measurements were repeated a third time. The average of the 2 Lange™ skinfold measurements within 1 mm for each site were manually averaged and summed together. Body density values were calculated using the generalized sum of 3 skinfold equation of Jackson and Pollock for the men (16). The body density values were then converted to %BF using the formula of Siri (25): %BF = [(4.95/body density) * 4.50] * 100. Lean body mass was calculated by subtracting fat mass from total body mass. Height (cm) and body mass (kg) were determined to the nearest 10th of a unit using a Detecto 439 (Detecto Scale Company, Webb City, MO, USA) physician scale.

Linear Bat Swing, Angular Hip, and Angular Shoulder Velocities

The lead author demonstrated proper bat swing technique and verbally explained the bat swing warm-up and hitting protocol to all participants. Each hitter performed stretches of their choice for 5 minutes before taking 2 sets of 10 warm-up swings with an 83.8 cm, 851 g (33 in, 30 oz) baseball bat. The stretching routine was duplicated on subsequent testing sessions.

After performing warm-up swings, participants had reflective markers placed on their body before stepping into the batting tee station. Thirteen markers were used in this study. Markers 1-4 were attached to the modified test bat. Markers 5-8 were attached to the upper body. Markers 9-13 were located on the hip region. The article by Szymanski et al. (33) details the location of the reflective markers on the bat and body. The height of the batting tee was adjusted to the height of the hitter's pubic arch, recorded, and replicated for reliability. This location represented a baseball pitched in the center of each hitter's strike zone. At the batting tee station, the hitter performed 10 total swings, 4 practice swings, and 6 recorded swings, with the modified test bat of the same length and mass used in previous research (30,33,34). The average velocity of the participant's final 6 swings recorded with a motion capture system (Motion Reality Inc., Marietta, GA, USA) represented their AHV, ASV, and linear BV. Test-retest reliability of these variables was r > 0.85, 0.88, and 0.95 (p ≤ 0.05). A detailed description of this protocol, instrumentation, reliability, and calibration can be found in Szymanski et al. (33).

Torso Rotational Strength

A Cybex Torso Rotation Machine™ (Cybex International, Medway, MA, USA) was used to measure 3RM dominant and NDTRS. Standing torso rotations (warm-up exercise) were performed before torso rotational strength testing. Participants sat in the Cybex Torso Rotation Machine™, putting both feet on the foot plates while squeezing their knees securely against adductor pads. The seat height was adjusted and recorded so that the participant's knees were at a 90° angle to the foot plates. Participants sat upright, grasped the handles with each hand, and pulled their chests firmly to the chest pads. A warm-up protocol similar to the 3RM PS and BP test (3) was used before participants attempted the 3RM torso rotational strength test. Participants performed a 3RM torso rotational strength test from their dominant side first. This was the direction they swung their bats. After completing the 3RM torso rotational strength test for the dominant side, the participant's nondominant side was measured. This allowed the lead author to evaluate unilateral rotation strength. Participants performed a torso rotation that was similar to the range of motion of a baseball swing (−30° to +75° = 105°) and similar to the range of motion previously reported (36). One of the limitations of the Cybex Torso Rotation Machine™ was that the participant's body (chest, hips, and thighs) could not be completely restrained from movement. Therefore, participants could have some forward, backward, or lateral movement of the shoulders, hips, or legs. Each participant's movements were monitored and corrected according to the directions of the Cybex Torso Rotation Machine™. If the participant failed to complete the full range of motion (105°) or did not perform the test properly, he was given 2 minutes of rest and then asked to perform another test with the load decreased by 2.3 kg. Test-retest reliability for both of these variables was r > 0.90 (p ≤ 0.05).

Sequential Hip-Torso-Arm Rotational Strength: Medicine Ball Hitter's Throw

A 1-kg, 2-handed medicine ball hitter's throw test for maximum distance was used to assess sequential hip-torso-arm rotational strength (33,36). A suspended 0.75-m square (target), through which the participant was required to throw the medicine ball, was positioned 3.0 m in front of the participant at a height of 0.75 m. Pilot work reported by Szymanski et al. (36) demonstrated a high statistically significant correlation between test-retest reliability performed by college baseball players on 2 consecutive days (r = 0.96, p < 0.001). A 1-kg medicine ball was used because it is approximately the same mass as the normal game bat (0.85 kg) of each participant. Before the 2 familiarization practice sessions began, during the first week of the study, the lead author demonstrated the MBHT test. Then the participants were allowed to perform as many practice throws as they desired until they were able to make 3 consecutive throws with correct mechanics within 0.50 m of their longest practice throw as determined by the principal investigator.

On the testing day, the participants were instructed to stand in their normal game batting stance, holding the medicine ball at their back shoulder height with 2 hands behind a white taped line. They were then asked to throw the medicine ball (similar to the movement of their normal batting swing) for maximum distance. The medicine ball throwing technique for each participant was monitored, and corrections were made to reinforce traits identified for swinging a baseball bat by Breen (6) and Race (18) as determined by the principal investigator. In an attempt to maximize reliability for each testing session, foot placement was recorded to assure that each movement was duplicated. Distance was measured from the front of the white line to the closest edge of the medicine ball imprint. The participant was given 2 practice throws to coordinate his aim through the square (target), followed by 3 maximal efforts. The best distance (in meters) was recorded.

Lower Body Power: Vertical Jump

A Vertec™ (Sports Imports, Columbus, OH, USA) apparatus was used to measure countermovement VJ height. Each participant stood flat footed with their feet shoulder width apart and raised their dominant hand straight up over their head. The participant then walked forward moving the vanes of the Vertec™ while a research assistant made sure that the participant's arm was fully extended. This was used to measure the base line standing reach. Each participant was then allowed 3 countermovement maximal VJs with the highest jump height recorded and used for data. Test-retest reliability for VJ was r > 0.88 (p ≤ 0.05). Recently, Duncan et al. (10) stated that researchers have disagreed as to which specific equation is best to estimate peak power in watts from a VJ test because results have varied when compared to actual peak power data measured from a force platform. Equations have been created for various athletes (7,22), genders (7,13,22), and jump techniques (7,13,22). Duncan et al. (10) used 25 elite junior male basketball players with a mean age of 16.5 years. Actual peak power and maximal countermovement VJ height were assessed using a force platform. Results indicated that the Sayers et al. (22) countermovement VJ equation most accurately estimated actual peak power compared to the Harman et al. (13) squat jump, Sayers et al. (22) squat jump, and Canavan and Vescovi (7) countermovement VJ estimate of peak power equations. Because the subjects and VJ technique used in the current study are similar to the ones used by Duncan et al. (10), we used the Sayers et al. (22) countermovement VJ equation to estimate peak power for the high-school baseball players. The Sayers et al. (22) countermovement VJ equation used to calculate estimated peak power is Peak Power (W) = 51.9 × VJ (cm) + 48.9 × BM (kg) - 2007. This equation was used to investigate the relationship of VJ estimated peak power to linear BV because if 2 players of different body mass both jumped the same height, their absolute VJ score would be the same regardless of their body mass. Because 1 of the 2 variables that determine VJ height is ground reaction force, the amount of force needed in a VJ depends on the subject's body mass. The greater the force-output-to-body-mass ratio, the greater the VJ height, and therefore, the greater the resulting power output. This equation allowed us to consider who was “pound for pound” more powerful.

Muscular Strength: Parallel Squat and Bench Press

According to Baechle et al. (3), many of the participants in this study would be classified as a beginner or intermediate lifter (<1 year of resistance training experience). Because of this training status, an estimation of 1RM (the most amount of weight lifted 1 time) was determined by performing 3RM tests (the most amount of weight lifted 3 times) on the PS and BP using Olympic standard free weights.

A regimen of full-body, dynamic warm-up exercises was performed before all testing and training sessions. There was 3 minutes of rest between near-maximal lifts (36). The 1RM was estimated using the load assessment table adapted from Wathen (39). The 3RM for PS and BP was assessed to estimate 1RM at 0 and after 4, 8 (to ensure that appropriate % was used during training), and 12 weeks of training using the methods described by Szymanski et al. (36). The 3RM was determined to be the maximal weight lifted after 2 consecutive unsuccessful trials. The progression of incremental load increases used for both tests had already been established for 1RM testing (3). Weight belts were worn during near-maximal PSs. Proper spotting techniques were demonstrated and used for all exercises. In an attempt to maximize reliability from pre to posttest, both the foot placement and squat depth were recorded and replicated using a Z-Squat. An adjustable bungee cord was stretched between 2 metal poles with holes marked every inch at a level that assured that the 90° of knee flexion was repeated. The 90° squat depth was determined when the upper thigh was parallel to the floor (knee is in 90° of flexion). When the buttocks of the participant touched the cord, the participant performed the concentric phase of the PS. Test-retest reliability for BP and PS were r > 0.97 and 0.94 (p ≤ 0.05). The lead author used a weight training percentage table to determine the appropriate resistance (%) of the predicted 1RM for PS and BP for each participant during training sessions (38).

Resistance Training Protocol

Resistance training for both groups was performed 3 d·wk−1 for 12 weeks during the off-season (fall semester) in the morning before classes began in an air-conditioning facility according to a stepwise periodized method similar to previous research (30,33,34). Two warm-up sets of 10 repetitions for the core strength exercises (PS and BP) were completed to prepare the participants before performing the more demanding 3 working sets. Furthermore, those participants in group 2 performed medicine ball exercises 3 d·wk−1 for 12 weeks. Set workloads were progressively increased every 4 weeks during the study after having 3RM PS and BP reassessed. Additionally, various assistance exercises (stiff-leg deadlift, dumbbell row, shoulder press, biceps curls, and triceps extensions) were performed to make the training more comprehensive and realistic to the off-season training programs of high-school baseball players. Furthermore, those participants in group 2 performed medicine ball exercises 3 d·wk−1 for 12 weeks. Specific rotational medicine ball exercises were chosen and performed 2 d·wk−1 for this study because they mimic the sequential, ballistic, and rotational movements of hitting and throwing a baseball (36). Other explosive, whole-body medicine exercises were performed 1 d·wk−1 when resisted leg exercises were not performed (36). Also, the mass of the medicine ball and number of repetitions progressed every 4 weeks from greater to lighter mass and fewer to greater repetitions to allow each participant to perform each exercise at velocities as close to “game speed” as possible. Training protocols and schedule of exercises are given in Tables 2 and 3.

Table 2
Table 2:
Training protocols.*†
Table 3
Table 3:
Schedule of exercises.

Statistical Analyses

SPSS (version 11.5; SPSS, Inc., Chicago, IL, USA) was used for the statistical analysis. The relation between various anthropometric and physiological variables to linear BV was determined using a Pearson's product moment correlation. Interpretations of magnitude and critical values of r regarding correlation coefficients were referenced by Safrit and Wood (21). Correlations were listed as high (±0.80 to 1.0), moderately high (±0.60 to 0.79), moderate (±0.40 to 0.59), or low (±0.20 to 0.39). The coefficient of determination, used for interpreting the meaningfulness of the relation, was developed by squaring the correlation, multiplying by 100, and expressing as a percentage.

Independent sample t-tests were conducted before the 12-week study to determine if any statistically significant differences existed between the 2 groups. To determine if any statistical differences existed between or within groups, 2 (group) × 2 (trials) repeated-measure analyses of variance were conducted on BV, AHV, ASV, VJ, peak power, MBHT, torso rotational strength, BP, and PS. To detect the specific treatment effect on linear BV, an independent samples t-test was done on the difference between pre and posttest scores (delta scores) for both groups. Effect size and observed power were also examined. Statistical significance was accepted at an alpha level of p ≤ 0.05. All data are reported as mean ± SD.


Independent sample t-tests indicated no statistically significant differences between groups for height, body mass, %BF, LBM, torso rotational strength, lower body power, upper and lower body muscular strength, MBHT, AHV, ASV, or BV before training. See Tables 1 and 4. Table 4 shows group 1 and 2 results of velocity, power, and strength data and percent (%) change for pre and posttraining data reanalyzed from Szymanski et al. (33). These data show what statistically changed over the 12-week resistance training program.

Table 4
Table 4:
Pre and posttraining mean (±SD) for velocity, power, and strength data for high-school baseball players, and percent (%) change.*

Table 5 shows the relationship among preanthropometric and physiological variables and prelinear BV for groups 1 and 2. Table 6 shows the relationship among postanthropometric and physiological variables and postlinear BV for groups 1 and 2.

Table 5
Table 5:
The relationship between various measures of preanthropometric, power, strength, and velocity to pre-BV for G1 and G2.*†‡
Table 6
Table 6:
The relationship between various measures of postanthropometric, power, strength, and velocity to post-BV for G1 and G2.*†‡


The objective of this study was to report significant relationships between anthropometry and physiological performance variables of high-school baseball players and linear BV before and after performing 12 weeks of resistance training. Significant relations were found between numerous variables and linear BV for both groups 1 and 2 pre and posttraining.

The analysis of the anthropometric test results indicated that LBM had a pretraining (r = 0.59 and 0.61) and posttraining (r = 0.64 and 0.68) moderately high relationship with linear BV for groups 1 and 2, whereas height and body mass moderately correlated with linear BV for both groups 1 and 2 pre and posttraining (see Tables 5 and 6). These findings are similar to other studies that investigated adolescent (29), high-school (19,38), and college (4,5,27,31,37) baseball players and suggests that high-school baseball players who are taller with longer levers and have greater LBM are able to generate more force at impact than players with shorter levers and less LBM. This may be important information for the college baseball coach who is recruiting high school players and for the major league baseball scout evaluating young players for the major league draft. When looking at the data in Tables 5 and 6, one will notice that the relationship of post-LBM to postlinear BV was larger than the same predata for both groups, suggesting that strength and conditioning coaches should strive to increase player's LBM during the off-season. Based on the extent of the coefficient of determination, 35 and 37% and 41 and 46% of the linear BV could be predicted by LBM for pre and posttraining for both groups 1 and 2. Interestingly, the relationship between the current high-school baseball player's LBM and body mass was highly correlated, with r values of 0.90 and 0.86 for pretraining and 0.92 and 0.88 for posttraining data for groups 1 and 2, respectively. The coefficient of determination for these relationships was 81 and 74% for pretraining and 85 and 77% for posttraining.

Szymanski et al. (32) recently indicated in a brief review that 11 studies have examined the relationship between strength (2,4,5,15,19,26-29,31,38) and power (2,5,27-29,31) to BV of adolescent, high-school, and college baseball players, and college students. The findings of the current study are in agreement with the studies discussed in the review article (32). The current data in this article found significantly high relations between post-NDTRS (r = 0.83) for group 1 and post-DTRS (r = 0.81) and NDTRS (r = 0.84) for group 2 and postlinear BV. This indicates that 69, 67, and 71% of linear BV could be predicted by posttraining torso rotational strength, respectively. Moderately high significant relations existed between pre-NDTRS (r = 0.66) for group 1 and pre-DTRS (r = 0.60) and NDTRS (r = 0.62) for group 2, and post-DTRS (r = 0.60) for group 1 and linear BV. This indicates that 44, 36, 38, and 36% of linear BV could be predicted by pre and posttraining torso rotational strength. Moderately high significant relations between pre and postestimated peak power (Sayers et al. [22] peak power countermovement jump) for groups 1 (r = 0.59 and 0.70) and 2 (r = 0.63 and 0.73) existed with linear BV. This indicates that 35, 49, 40, and 53% of linear BV could be predicted by estimated lower body power. Significant moderately high relationships existed between post-upper body (r = 0.65) and lower body (r = 0.62) strength and postlinear BV for group 1. This indicates that 42 and 38% of linear BV could be predicted by BP and PS strength. Significantly moderate relationships existed between pre-upper body (r = 0.56 and 0.45) and prelower body (r = 0.51 and 0.46) strength and prelinear BV for groups 1 and 2, pre-MBHT (r = 0.42) for group 2, and post--upper body (r = 0.65 and 0.56) and lower body (r = 0.62 and 0.46) strength for groups 1 and 2, and post-MBHT (r = 0.49 and 0.59) for groups 1 and 2 and linear BV of high-school baseball players. This indicates that 31, 20, 26, 21, 42, 31, 38, and 21% of pre and postlinear BV could be predicted by pre and post-BP and PS strength, and 18 and 35% of pre and postlinear BV for group 2, and 24% of postlinear BV for group 1 could be predicted by rotational power. Significant low relationships existed between pre-VJ (r = 0.30 and 0.36) for both groups, and pre- and post-AHV (r = 0.39 and 0.36) for group 2 and prelinear BV. When looking at Tables 5 and 6, one will see that all of the poststrength and power relationships to postlinear BV were greater than predata for the same variables, except post-PS to postlinear BV. This relationship remained the same. These data suggest that strength and conditioning coaches should strive to improve strength and power of baseball players during the off-season to improve linear BV. Grip strength of high-school baseball players has also demonstrated significant relationships to BV (26,27,29); however, grip strength was not measured with the current players. Researchers who have studied college baseball players have also found that those with the greater lower body strength (4), upper body strength (20), lower body power (27,37), rotational strength (28), and rotational power (5,28,31,37) have greater BV. Researchers (5,27,31) also suggest that college baseball players with greater grip strength have greater BV. These data (4,5,19,26-29,31,37,38), in agreement with the current study, collectively indicate that baseball players with greater BV are strong and powerful individuals who are taller and have a greater LBM.

To our knowledge, this is the first study evaluating estimated peak power from VJ height to linear BV of high-school baseball players. The reason for this was that even though VJ height measured during posttesting significantly increased (6.0 and 5.7%) and had a moderate correlation (r = 0.43 and 0.60) to linear BV for groups 1 and 2, if one considers the body mass of players and looks at the linear BV raw data, some players who jumped the same height had greater body mass had greater linear BV. Therefore, the Sayers et al. (22) countermovement equation used to estimate peak power from VJ height was used. This data indicates that there is a posttraining moderately high correlation (r = 0.70 and 0.73) for groups 1 and 2 between those who are more powerful from estimating peak power and those that swing the baseball bat with greater velocity. Hence, we recommend using the Sayers et al. (22) countermovement estimated peak power equation with VJ height data to further evaluate high-school baseball player's performance variables if the same VJ protocol and similarly aged baseball players are being evaluated. On a side note, we also used the other 3 equations for estimating peak power mentioned earlier in this article, and they all produced moderately high correlations to linear BV even though some of them used a squat jump as opposed to a countermovement jump to estimate peak power (see Tables 5 and 6).

This research provides a foundation for the development of baseball-specific training programs, and information that can be used for recruiting and/or scouting. The research design of this study was to investigate correlations, which must be interpreted with caution because relationships do not imply causality. Table 4 lists the pre and posttraining results of the 2 groups in the current study. When reading the data results, one will notice that linear BV and all of the other power and strength variables significantly increased for both high school groups after 12 weeks of training. Some researchers (5,20,24,26,27,31,38,40) have suggested that those that strength train baseball players should focus on developing lower body power, and upper and lower body strength to attempt to increase BV. The results of this study do indicate that group 1, which completed a 12-week stepwise periodized resistance training program, made significant increases in linear BV (3.6%), but not as great as group 2 (6.0%). These data support previous research that suggests that strength and power be improved in the weight room to improve a baseball player's BV. However, group 2, which performed the same resistance training program while completing additional medicine ball exercises, improved significantly more. This means that although power and upper and lower body strength did have a significant relationship to increases in linear BV, both pre and posttraining, it does not mean that those posttraining improvements collectively caused the direct posttraining improvements in linear BV. Szymanski et al. (33,34) reported that improvements in traditional strength training variables (upper and lower body strength, or wrist and forearm strength) do not directly lead to improvements in linear BV. A previous article suggested that the greater improvement in linear BV of group 2 (treatment group) was because of improvement of torso rotational strength and AHV that were developed by the implementation of an additional rotational medicine ball program that essentially mimicked the actions of swinging a baseball bat, thus supporting the concept of sport specificity (33). When one looks at college baseball players, research findings indicated that leg power (27,31) and rotational power (5,28,31) were significantly related to BV. However, results from a most recent study (35) that reported the physiological and anthropometric characteristics of college baseball players over an entire year with the same subjects and found that strength variables (squat, BP, and 1-arm dumbbell row) and leg power improved from pretraining to preseason, but BV did not change. This indicates that more mature college baseball players do not improve BV by performing resistance training like previous research has demonstrated with high school players (30,33,34). Thus, strength coaches should be careful not to suggest that simply improving power and upper and lower body strength will have the same effect on all levels of baseball players (high school, college, and professional). Maturation, initial strength, weight training experience, specificity of training, and the biomechanics of hitting must be considered. Therefore, we suggest that the results presented in this article and the other correlation studies mentioned in this article only be used to identify and understand general relationships between anthropometric and physiological variables and linear BV. This does not mean that resistance training that develops strength and power is not important for the baseball player. It means that resistance training only plays a part in the complete development of a baseball player, especially as it relates to BV. Recent research findings indicated that strength and power, and BV, were maintained over the entire competitive season (35). Resistance training provided an opportunity for these players to play optimally on the field because they stayed healthy, strong, and powerful.

Practical Applications

The results of this article indicate that although there are statistically significant relations between various anthropometric and physiological variables to linear BV, there is also considerable variation within the factors that contribute to linear BV of high-school baseball players. This may indicate that, to a certain extent, there may need to be specific resistance training strategies implemented to improve linear BV, especially for more mature, experienced baseball players.

The data in this article, in agreement with previous research, suggest that a strength training program designed to improve linear BV for high-school baseball players should have goals to improve LBM, lower body power, lower and upper body strength, rotational strength, and explosive rotational power. The 12-week stepwise resistance training program outlined in this article and used in previous research (30,33,34) has accomplished these goals for high-school baseball players.

For those who want to incorporate more baseball-specific training strategies, the dry swing or batting practice over and underweighted (±12% standard bat weight) bat program by DeRenne et al. (8) that increased BV by 6 and 10% after 12 weeks of training could be implemented. Dry swings (taking swings without hitting a baseball) with over and underweighted bats could be taken in the weight room (if there is enough room to swing safely) or batting cage area during baseball practice, or players could take batting practice with over and underweighted bats. This would be a natural part of baseball batting practice and may be more motivational because players will probably consistently swing harder to see how hard and/or far they can hit the baseball. Furthermore, it must be mentioned that if one wants to have greater BV, one should develop outstanding bat swing mechanics and/or swing a bat with a smaller moment of inertia (1,6,11,18). This simply means a bat that is shorter and/or lighter.

Finally, the tests used to collect data about the baseball players in this article were selected because they provided information regarding the overall strength, power, and physiques of teen-aged high-school baseball players. In agreement with previous research (14,37), the use of these tests in player selection may provide valuable information to those that recruit players for college, or draft or sign players for professional baseball.


1. Adair, RK. The Physics of Baseball (2nd ed.). New York, NY: Harper & Row, 1994.
2. Albert, JM, Szymanski, DJ, and Stanley, BE. Relationship between changes in physiological characteristics and softball-specific variables of NCAA Division I softball players. J Strength Cond Res 22: 83-84, 2008.
3. Baechle, TR, Earle, RW, and Wathen, D. Resistance training. In: Essentials of Strength and Conditioning. Baechle, TR and Earle, RW, eds. Champaign, IL: Human Kinetics, 2000. pp. 395-425.
4. Basile, R, Otto, RM, and Wygand, JW. The relationship between physical and physiological performance measures and baseball performance measures. Med Sci Sports Exerc 39: S214, 2007.
5. Bonnette, R, Spaniol, F, Melrose, D, Ocker, L, Paluseo, J, and Szymanski, D. The relationship between rotational power, bat speed, and batted-ball velocity of NCAA Division I baseball players. J Strength Cond Res 22: 112-113, 2008.
6. Breen, JL. What makes a good hitter? JOHPER 38: 36-39, 1967.
7. Canavan, PK and Vescovi, JD. Evaluation of power prediction equations: Peak vertical jumping power in women. Med Sci Sport Exerc 36: 1589-1593, 2004.
8. DeRenne, C, Buxton, BP, Hetzler, RK, and Ho, KW. Effects of weighted bat implement training on bat swing velocity. J Strength Cond Res 9: 247-250, 1995.
9. DeRenne, C and Okasaki, E. Increasing bat velocity (Part 2). Athletic J February: 54-55, 1983.
10. Duncan, MJ, Lyons, M, and Nevill, AM. Evaluation of peak power prediction equations in male basketball players. J Strength Cond Res 22: 1379-1381, 2008.
11. Escamilla, RF, Fleisig, GS, DeRenne, C, Taylor, MK, Moorman, CT, and Andrews, JR. A comparison of age on baseball hitting kinematics. J Appl Biomech 25: 210-218, 2009.
12. Escamilla, RF, Fleisig, GS, DeRenne, C, Taylor, MK, Moorman, CT, Imamura, R, Barakatt, E, and Andrews, JR. Effects of bat grip on baseball hitting kinematics. J Appl Biomech 25: 203-209, 2009.
13. Harman, EA, Rosenstein, MT, Frykman, PN, Rosenstein, RM, and Kraemer, WJ. Estimation of human power output from vertical jump. J Appl Sport Sci Res 5: 116-120, 1991.
14. Hoffman, JR, Vazquez, J, Pichardo, N, and Tenenbaum, G. Anthropometric and performance comparisons in professional baseball players. J Strength Cond Res 23: 2173-2178, 2009.
15. Hughes, SS, Lyons, BC, and Mayo, JJ. Effect of grip strength and grip strengthening exercises on instantaneous bat velocity of collegiate baseball players. J Strength Cond Res 18: 298-301, 2004.
16. Jackson, AS and Pollock, ML. Generalized equations for predicting body density of men. Br J Nutr 40: 497-504, 1978.
17. Jackson, AS and Pollock, ML. Practical assessment of body composition. Phys Sports Med 13: 76-90, 1985.
18. Race, DE. A cinematographic and mechanical analysis of the external movements involved in hitting a baseball effectively. Res Q 32: 394-404, 1961.
19. Reed, JG, Szymanski, DJ, Albert, JM, Hawthorne, LZ, Hemperley, DL, Hsu, HS, Skinner, CJ, and Tatum, JR. Relationship between physiological performance variables and baseball/softball specific variables of novice college students. J Strength Cond Res 22: 111-112, 2008.
20. Reyes, GF and Dolny, D. Acute effects of various weighted bat warm-up protocols on bat velocity. J Strength Cond Res 23: 2114-2118, 2009.
21. Safrit, MJ and Wood, TM. Basic Statistics. In: Introduction to Measurement in Physical Education and Exercise Science (3rd ed.). St. Louis, MO: McGraw-Hill, 1995. pp. 43-84.
22. Sayers, SP, Harackiewicz, DV, Harman, EA, Frykman, PN, and Rosenstein, MT. Cross-validation of three jump power equations. Med Sci Sport Exerc 31: 572-577, 1999.
23. Sergo, C and Boatwright, D. Training methods using various weighted bats and the effects on bat velocity. J Strength Cond Res 7: 115-117, 1993.
24. Shaffer, B, Jobe, FW, Pink, M, and Perry, J. Baseball batting: An electromyographic study. Clin Orthop Rel Res 292: 285-293, 1993.
25. Siri, WE. Body composition from fluid spaces and density. In: Techniques for Measuring Body Composition. Brozek, J and Hanschel, A, eds. Washington, DC: National Academy of Sciences, 1961. pp. 223-244.
26. Spaniol, F, Bonnette, R, and Melrose, D. The relationship between grip strength and bat speed of adolescent baseball players. J Strength Cond Res 20: 747, 2007.
27. Spaniol, F, Bonnette, R, Melrose, D, and Bohling, M. Physiological predictors of bat speed and batted-ball velocity in NCAA Division I baseball players. J Strength Cond Res 20: 185, 2006.
28. Spaniol, F, Paluseo, J, Bonnette, R, Melrose, D, Ocker, L, and Szymanski, D. The relationship between static strength, rotational strength, rotational power, bat speed, and batted-ball velocity of NCAA Division I baseball players. J Strength Cond Res 24(Suppl. 1): 1, 2010.
29. Spaniol, FJ. Physiological predictors of bat speed and throwing velocity in adolescent baseball players. J Strength Cond Res 16: 16, 2002.
30. Szymanski, DJ, Albert, JM, Hemperley, DL, Hsu, HS, Moore, RM, Potts, JD, Reed, JG, Turner, JE, Walker, JP, and Winstead, RC. Effect of overweighted forearm training on bat swing and batted-ball velocities of high school baseball players. J Strength Cond Res 22: 109-110, 2008.
31. Szymanski, DJ, Albert, JM, Reed, JG, and Szymanski, JM. Physiological predictors of sport-specific skills of Division I collegiate baseball players. In: Proceedings of the 36th Annual Meeting of the SEACSM. Torok, D, ed. Birmingham, AL: Southeastern Chapter of the American College of Sports Medicine, 2008. P27. pp. 34.
32. Szymanski, DJ, DeRenne, C, and Spaniol, FJ. Contributing factors for increased bat swing velocity. J Strength Cond Res 23: 1338-1352, 2009.
33. Szymanski, DJ, McIntyre, JS, Szymanski, JM, Bradford, TJ, Schade, RL, Madsen, NH, and Pascoe, DD. Effect of 12 weeks of torso rotational strength on angular hip, angular shoulder, and linear bat swing velocities of high school baseball players. J Strength Cond Res 21: 1117-1125, 2007.
34. Szymanski, DJ, McIntyre, JS, Szymanski, JM, Molloy, JM, Madsen, NH, and Pascoe, DD. Effect of wrist and forearm training on linear bat-end, center of percussion, and hand velocities, and time to contact of high school baseball players. J Strength Cond Res 20: 231-240, 2006.
35. Szymanski, DJ, Szymanski, JM, Albert, JM, Beam, JR, Hsu, HS, Reed, JG, and Spaniol, FJ. Physiological and anthropometric characteristics of college baseball players over an entire year. J Strength Cond Res 24(Suppl. 1): 1, 2010.
36. Szymanski, DJ, Szymanski, JM, Bradford, TJ, Schade, RL, and Pascoe, DD. Effect of 12 weeks of medicine ball training on high school baseball players. J Strength Cond Res 21: 894-901, 2007.
37. Szymanski, DJ, Szymanski, JM, Brooks, KA, Braswell, MT, Britt, AT, Hsu, HS, Lowe, HE, Taylor, EG, and Weil, KL. The relationship between power and lean body mass to sport-specific skills of college baseball players. Med Sci Sports Exerc 41: 307-308, 2009.
38. Szymanski, JM, Szymanski, DJ, Albert, JM, Hemperley, DL, Hsu, HS, Moore, RM, Potts, JD, Reed, JG, Turner, JE, Walker, JP, and Winstead, RC. Relationship between physiological characteristics and baseball-specific variables of high school baseball players. J Strength Cond Res 21: 747, 2008.
39. Wathen, D. Load assessment. In: Essentials of Strength Training and Conditioning. Baechle, TR, ed. Champaign, IL: Human Kinetics, 1994. pp. 435-439.
40. Welch, CM, Banks, SA, Cook, FF, and Draovitch, P. Hitting a baseball: A biomechanical description. J Orthop Sports Phys Ther 22: 193-201, 1995.

bat speed; correlations; performance variables

© 2010 National Strength and Conditioning Association