Throwing arm injuries afflict baseball pitchers at all levels of competition (youth through to professional). Epidemiological evidence of Major League Baseball injuries suggest that professional pitchers incur about 34% higher injury incidence rates compared with fielders. The upper extremity accounts for 51.4% of those injuries, with 46–62% of the days on the disabled list attributable to arm injuries (1,14).
Lyman et al. (9,10) first reported a positive relationship between the number of pitches thrown and arm pain in youth pitchers. Youth pitchers who self-reported the number of innings pitched indicated that throwing over 100 innings per season increased injury risks 3.5 times (7). New evidence suggest that multiple risk factors increase susceptibility of pitching-arm injuries, which include (a) pitching more than 8 months per year, (b) pitching more than 100 innings per year, (c) throwing more than 80 pitches per game, (d) pitching at velocities greater than 85 mph, and (e) the frequency of sliders and curveballs pitched (6,7,9,10,13,14). It is currently believed that arm fatigue exacerbates risk from these factors. Among adolescents who responded to a pitching survey, Olsen et al. (13) found those who pitch regularly, despite arm fatigue, are 36 times at greater risk of injury.
To lessen the risk of injury among youth and amateur participants, Baseball USA and Little League Baseball instituted age-related pitch counts and rest days guidelines to help prevent pitching injuries (6), whereas 100 pitches per game is the limit set for professional baseball with many throwing more than this workload designation. In addition, Major League Baseball has imposed workload guidelines premised on limiting the innings pitched per season among young professional pitchers, otherwise known as the “Verducci Effect ” Although not unanimously supported, the Verducci Effect provides a negative prognosis of workload, whereby young pitchers who substantially increase the number of innings pitched over subsequent seasons are at greater risk of injury (15). Pitcher abuse points are accrued, where accumulation beyond 100 pitches per game is considered contraindicative (15). Despite the implementation and monitoring of these metrics, injury statistics have not declined, as pitchers continue to throw more than 100 pitchers per game and in some cases consecutively (1,14).
Biomechanical studies on extended pitching (greater than 6 innings pitched or 100 pitches thrown) have shown ball velocity to be a quantitative measure impacted by arm fatigue, where starting pitchers' FBVs have been shown to decrease 2–5 mph from the first inning to the last (12). In the study by Escamilla et al. (5), a 2.2-mph decrease in ball velocity was observed between the first 2 and last 2 innings of simulated play, under laboratory settings. Recent evidence depict a 2-mph drop in ball velocity within a simulated game, which Gandhi et al. (8) attributed to reduced infraspinatus involvement with concomitant external rotation strength deficits. Empirical evidence supports the use of velocity decrements as a surrogate measure of arm fatigue, yet a paucity of literature exists in reporting pitchers' FBVs between games or during a competitive season. Although tracking and charting game ball velocities is commonly practiced, analysis of ball velocity trends across games have not been explored, where it remains unknown whether professional pitchers experience significant velocity changes throughout the competitive season.
By analyzing mean FBV trends of minor league pitchers during the course of a competitive season, the purpose of our retrospective study was to (a) assess the FBV response between games, (b) indicate the relationship between the numbers of rest days and FBV differences during the subsequent outing for a pitcher, and (c) to determine whether the ratio of pitcher work to rest (number of pitches thrown previously to days' rest) relates to subsequent day FBV changes. We propose FBV decreases as the season progresses, much in the same manner that ball velocity can decrease within a single game. We assume that accumulation of pitches would cause similar velocity decreases within-games producing velocity decreases between-games pitched. We also hypothesize that greater number of days rest before a game pitched would minimize velocity decreases and that higher ratios of pitches to rest days (more pitches with less rest) would elicit greater velocity reductions.
Experimental Approach to the Problem
A retrospective analysis of pitching records among minor league baseball pitchers was performed to identify velocity trends during the competitive season, address game workloads, appearances, and rest sequences. Competitive pitching records represent true measures of velocity that professional pitchers incur in a competitive environment. These records can be made available to the strength and conditioning professional to chart and assess with respect to training demands, timing of training, and athletes' pregame preparation habits. Our analyses focused on potential changes to mean FBVs throughout a minor league season, where the number of rest days formulated a pitching work to rest ratio (PWRR), which could be used to determine whether a pitcher increased or decreased his FBV in his next game.
The pitching records of 12 minor league pitchers who competed during the 2009 Class-A Short Season were analyzed. In total, 9 right-handed and 3 left-handed male pitchers were involved (mean age 21.2 ± 1.1 years, height 1.87 ± 0.048 m, mass 86.9 ± 8.51 kg). All pitchers were healthy without any prior reported injuries. Baseball experience was estimated from the age of 9 years old, being approximately 10–12 years. Pitching experience could not easily be determined from the sample population, as some players were converted from position players to pitchers in high school and college. Inclusion required pitchers to play in at least 8 games over the minor league season, with entry into games at the beginning, midpoint, or final 2 innings. It should be noted that most Class-A Short Season teams institute a “piggy back system” for starting pitchers, where approximately 3.5 innings are allotted for 2 starting pitchers. Relievers generally throw 1–2 innings each. Pitchers who did not complete the full 2009 Class-A Short Season were excluded from the study. The University at Buffalo's Institutional Review Board granted approval to conduct the study and pitching records were provided by St Louis Cardinals Major League Baseball affiliate.
Pitching Chart Analytics
Ball velocity and pitch type were tracked and recorded for all pitchers in every game played throughout the season. Velocities were obtained using a radar gun (Stalker Radar, Plano, TX) positioned directly behind home plate, accurate to within ±0.5 mph. On completion of the game, pitch data were compiled to include total number of pitches, type thrown, and inning counts for every pitcher. Altogether, 5743 pitches were recorded and indexed over the course of the season.
Mean game fastball velocities (GFBV) were derived for pitchers specific to each game they played and represent the mean velocity of fastballs thrown particular to that game. In addition, game pitches thrown (GPT) and game innings pitched (GIP) were tabulated for each pitcher for each game. Data from the first 8 games each pitcher threw were analyzed, given 8 represents the minimum number of games all pitchers played.
Pitchers' data were grouped and the mean FBV was computed for each game, i.e., for game 1 FBV denotes the mean of all GFBV of pitchers appearing in their respective first game. Similar determinations were derived for GPT and GIP to indicate average workload metrics over 8 games.
Regression analysis was performed using ordinary least squares, where the explanatory variable was game number (considered a measure of accumulation) and the dependent variable was FBV. A criterion of p < 0.05 was used to reject the null hypothesis. Additional analysis focused on whether velocity changes pitchers experienced later in the season could be explained by the number of rest days between games. Velocity differences (VDiff) were identified as changes in GFBV between consecutive games for a pitcher, where a positive VDiff signifies increased velocity between games and a negative VDiff denotes a reduction. Rest days were considered the cumulative number of days off between 2 successive games. For example, a pitcher who played on Monday and Thursday was deemed to have had 2 complete days of rest. The number of days rest and corresponding VDiff were treated as independent events and grouped together over all games for all pitchers. For pitchers who exceeded 8 days rest, they were included with the 8 days rest group. As a result, analysis included up to an 8 day rest interval, whereby the eighth day included days rest greater than or equal to 8. This relationship was then plotted with days rest as the independent variable and VDiff as the dependent variable to identify trends.
Last, we assessed the relationship between previous game pitch totals to rest days and the velocity changes pitchers experienced with subsequent outings. To account for any relationship between days rest and previous game pitch totals, we established a PWRR; determined by the ratio between the number of pitches thrown in the previous game and number of rest days between outings. Pitchers throwing a large number of pitches with low number of rest days incur high PWRRs as opposed to the same pitch count with more rest days. To examine the relationship between PWRR and VDiff, the PWRR was “binned” in equal intervals of 3, from 0 to 30 (30 being the maximum PWRR observed in the data). The VDiff for each instance in a given bin was then averaged. We treated PWRR and VDiff for each game independently across pitchers and across games, then plotted this relationship to identify trends.
“Rest days” from pitching included other activities. Most often, starting pitchers would not throw the day after their scheduled appearance, being day 2, where days 3, 4, and 5 involved long-toss and flat ground pitching. Day 3 was considered a sideline bullpen day consisting of 30–40 pitches. Relievers threw everyday based on feel, where intensities and distances were decided on by both the pitching coach and the pitcher. Relievers participated in strength and conditioning programming 4–5 times per week that featured total body exercises. Starting pitchers trained lower body on the day after their starts (day 2), with upper body on the day after their sideline bullpens (day 4). Relievers sprinted distances between 10 and 60 yards, whereas starters sprinted distances between 10 and 160 yards.
Mean FBV had shown a significant linear increase within the first 8 games of the season (R2 = 0.91, F(1,7) = 64.67, p < 0.001; Figure 1). The mean FBV presented with the SEM had significantly increased 0.25 (0.20) m/s (0.56 [0.44] mph) between the first game and the eighth game of the season; with the greatest significant velocity increase observed between the first and eighth game at 1.97 m/s (4.4 mph). The median FBV increased as well between the first and last games of the analysis from 38.6 m/s (86.4 mph) to 39.0 m/s (87.2 mph), respectively. This increase was not statistically significant.
Rest days between successive starts and the VDiff experienced in the subsequent game presented inconclusive results. Pitchers that had 3, 5, and 8 (or greater) days of rest returned in their next games with nonsignificant differences in mean velocities 0.46 m/s (1.02 mph), 0.21 m/s (0.46 mph), and 0.30 m/s (0.68 mph), respectively (Figure 2). However, pitchers with 2 and 6 days rest experienced a nonsignificant decrease in mean velocity of 0.42 m/s (0.95 mph) and 0.57 m/s (1.27 mph), respectively. The mean (SD) days rest experienced by the pitchers was 4.33 (3.86) days (Table 1), with 4 being the median and mode days rest observed during the competitive season.
PWRRs between successive starts and the VDiff displayed in subsequent games were also inconclusive (Figure 3). Pitchers that threw in a game with a PWRR of 3.1–6 and 27.1–30 showed a nonsignificant increase in mean velocity of 0.24 m/s (0.54 mph) and 0.45 m/s (1.01 mph), respectively. Meanwhile, pitchers competing with a PWRR between 15.1 and 18 responded with a nonsignificant decrease in mean velocity of 0.24 m/s (0.53 mph). Overall, no significant relationships were observed between the variables and no further statistical analyses were completed.
Contrary to our expectation, this study demonstrates that mean FBVs within a multistarter minor league system increases linearly during a season. Regression analysis predicted that any subsequent game thrown by a pitcher resulted in a significant 0.04 m/s (0.094 mph) increase in velocity, with mean FBV significantly increasing 0.25 m/s (0.56 mph) between the first and eighth game of the season. We also found that the number of days rest showed no relationship to the velocity difference exhibited in subsequent games. Likewise, the PWRR showed no relationship to VDiff in subsequent games.
The reason for the unexpected linear increase in FBV is uncertain. We hypothesized that mean FBVs would decrease during the 8-game period as a result of “fatigue” from the accumulated pitches thrown and innings pitched during the season. Our findings show that this is not the case in this sample of minor league pitchers. One explanation may be that fatigue within a season does not necessarily manifest itself as a velocity decrease. Neuromuscular and range of motion changes have been associated with higher pitch accumulation, both of which have been attributed to kinematic changes observed about the trunk, lead knee, and throwing shoulder (5,8,11,12). Biomechanical compensations are mechanical adaptations that may maintain performance despite fatigue. In baseball, pitchers impart biomechanical compensations to maintain peak velocity in response to increased exertion (5,12). Recently, stride length compensations indicated that pitchers can maintain peak ball and throwing hand velocities despite considerable changes in lower-body power output (4). As a result it is possible that our sample of pitchers were able to minimize expected decreases in velocity during the course of the season through performance maintaining adaptations.
It is possible that our sample of Class A Short Season baseball pitchers did not fatigue during the season. Perhaps the amount of rest allocated and activities between games at this level of play were sufficient, matched to GPT and GIP in maintaining performance. On average, pitchers threw 45.4 (±22.3) pitches per game, appeared in 3.15 (±1.52) innings per game, and had 4.33 (±3.86) days rest between appearances. The PWRR before each appearance was 36.0 (±38.7). More research is warranted to examine these workload standards in minor league baseball. Given our analysis of rest days and the PWRRs, we cannot substantiate or refute the claim that workload was matched by rest in Class A Short Season baseball.
Strength and conditioning practices may have adequately met the restoration of neuromuscular function, remediation of changes in joint ranges of motion, which may have reduced the onset of overtraining. Strength and conditioning platforms incorporated training intensities from 45 to 87% repetition maximum, with very little emphasis on muscular endurance in both weight training and conditioning. Volumes were kept low between 3 and 5 sets of no more than 6 repetitions per set to promote force and contractile velocity. Performance enhancing drugs may have been used in the off-season, where testing may not have been available, which could have affected velocity outcomes and is a limitation of our study. In-season supplement use was routinely monitored, where athletes could only consume products sanctioned by Major League Baseball's safety standards. It is possible that anabolic agents and stimulants could have been used off-premises, but this occurrence is highly unlikely.
Pitchers in this study may have experienced seasonal fatigue without fatigue manifesting as velocity decrease. Masked by confounding influences, our sample of pitchers may have improved skill factors (mechanics, psychological factors, training, nutrition counseling, etc) to reduce exertion-related velocity decline. If pitchers were not fatigued, it may be likely that the magnitude of velocity increases would be evident. Furthermore, pitchers may have conserved energy early within the competitive season, where increased effort was provided as the season developed. In this event, velocity increases may be considered injurious because greater physical effort under accumulation could further microdamage and biomechanical inconsistencies. This explanation is conjecture but warrants further investigation and attention to strength and conditioning practices to promote orthopedic health when late-season velocities escalate.
All competitive baseball organizations implement various protection methods to lessen pitcher injury risk, which include managing pitch counts, innings, and game appearances because cumulative repetitive stress is known to negatively influence long-term participation. In professional baseball, it is not uncommon to see a pitcher relieved during a game solely on ball velocity decrease. An intuitive extension of this logic could be able to manage long-term pitcher health based on FBV measurements (i.e., whether to allocate greater rest periods between games for relief pitchers with significant decline) rather than restrictions to pitches thrown and innings pitched. This study shows that although velocity screening is valid logic, it does not manifest itself as predicted by games pitched, pitches thrown, or days rest.
It remains possible that velocity decrements can be readily observed in injured pitchers or in pitchers approaching the point of substantial injury. All pitchers involved in this study were healthy and able to return to their previous game's pitching intensity, some even after as little as 1 day rest. What FBV trends occur in pitchers approaching the instance of catastrophic injury remains unknown. Therefore, velocity screening could prove to still have some merit in identifying the onset of severe injury in pitchers, close to, or before the time they themselves report it to a trainer or coach. Similarly, at levels where the pitching workloads are meticulously managed, such as the Short Season Class A level, strength and conditioning professionals have important roles in communicating with athletes, as velocity changes may not be discerned.
As the professional pitcher advances, workloads are extended to where pitchers may consecutively throw more than 100 pitches per game. The ability to assess competitive fatigue through FBV characteristics can aid baseball strength and conditioning professionals in adjusting training frequencies, timing, and workloads to promote throwing arm recovery and neuromuscular restoration of strength and range of motion. Baseball strength and conditioning professionals should concentrate in-season programming on undulated training patterns to promote recovery and maintain muscular power, both of which attempt to establish velocity consistency during competition. Essentially, the strength and conditioning professionals in baseball has greater control in managing physical stimuli within the professional baseball season, as game competition does not influence strength and conditioning practices. Exercise prescriptions should coincide with the time of the season, athletes' injury histories, and their competitive characteristics (pitch types, pitching role, and personal throwing programs, etc).
The athletes examined in this study were trained between 10 AM and 430 PM everyday with Sundays off from training for relief pitchers and starting position players. Exercise modalities and intensities were varied, as were the eccentric demands on the throwing arm in traditional weight training. It is recommended that in the event that velocity changes are noted (both increases and decreases above 2 mph), the strength and conditioning coordinator should dialogue with his/her athletes to determine if anatomical regions of the body have experienced greater fatigue, new soreness patterns, or instability. A decrease does not clearly delineate detraining, weakness, or injury, as velocity declines may be considered protective. Similarly, late-game velocity increases may not be considered advantageous, as greater risk of injury ensues as effort is amplified under a state of fatigue.
As an example, if a pitcher's mean velocity increases by 2 mph from previous outings without him purposely adjusting the biomechanics of his delivery, the strength and conditioning coach should be alerted. The same pitcher may coincidentally report a change in postgame pitching soreness patterns, indicating that the delayed onset of inflammation shifts from the latissmus dorsi of the throwing arm to his throwing biceps and flexor pronator mass. The strength and conditioning coordinator should attempt to manage exercise prescriptions to improve his/her recovery and restore muscle function.
Typically, starting pitchers organize their most intense training days on the days after their pitched games. The previously mentioned case warrants the strength and conditioning coordinator to improve rest and initiate training in the evening of the regularly scheduled training day or potentially wait 48 hours to initiate training. Exercise selections should be examined to improve sensitive anatomical regions' abilities to withstand elevated stress. If soreness is amplified in the flexor pronator mass and biceps, protective training exercises can be introduced. Examples from Crotin and Ramsey (2,3) can be integrated to improve medial elbow stability. Adjustment to exercises may be important in the event that a new biomechanical change has been instituted by the pitching coach. Consequently, continuous dialogue between the pitching coach and strength and conditioning coordinator can help protect pitchers from injury and improve acclimation to new mechanics throughout the season.
Our research indicates that short-season minor league pitchers have the capacity to return to their previous games' fastball intensities with varying degrees of rest. This indicates that no clear influence could be observed from traditional workload metrics such as previous pitch counts, game appearances, or innings pitched. Although not completely proven in this study, simply increasing the amount of days of rest seems to show no observable benefit for maintenance of FBV in Short Season Class A baseball, where innings pitched are restricted. This provides new insight toward reverting back to 4-man pitching rotations, where starting pitchers can throw with greater frequency, allowing greater involvement from 2 to 3 relief pitchers per game. This would be identified as a workload sharing model that differs from what is accepted by current professional baseball standards at the major league level.
Seasonal velocity increases may be an important indicator revealing the efficacy of an organization's workload management strategy in protecting pitching prospects. Recent practices in Major League Baseball involve early season shutdowns for young high-priced pitchers without empirical investigation, where workload tolerances based on innings pitched are assumed rather than evaluating pitches thrown and velocity characteristics. The goals of future works conducted on minor league pitching rotations should involve examinations of player workloads and training strategies to further enhance durability, career longevity, and enhancement of velocity. Collectively, such research can assist athletes in avoiding in-season shutdowns and injury-plagued participation at the major league level.
The research team would like to thank the St Louis Cardinals, Major League Baseball for their authorized use of professional baseball pitching charts. We also extend our gratitude to our research associates; Jennifer Martins, Philip Mathew, Joseph Westlake, Alyssa Herman, Connor McNally, and Laura Dipasquale for compiling the pitching chart data.
1. Conte S, Requa R, Garrick J. Disability days in major league baseball. Am J Sports Med 29: 431, 2001.
2. Crotin RL, Ramsey DK. Injury prevention for throwing athletes part I: Baseball bat training to enhance medial elbow dynamic stability. Strength Cond J 34: 79, 2012.
3. Crotin RL, Ramsey DK. Injury prevention for throwing athletes part II: Critical instant training. Strength Cond J 34: 49–57, 2012.
4. Crotin RL, Ramsey DK. Stride length compensations and their impacts on brace-transfer ground forces in baseball pitchers. In: Proceedings of the 36th Annual American Society of Biomechanics. American Society of Biomechanics, 2012. pp. 243–244. Available at: http://www.asbweb.org/conferences/2012/abstracts/357.pdf
. Accessed August 5, 2012.
5. Escamilla RF, Barrentine SW, Fleisig GS, Zheng N, Takada Y, Kingsley D, Andrews JR. Pitching biomechanics as a pitcher approaches muscular fatigue during a simulated baseball game. Am J Sports Med 35: 23, 2007.
6. Fleisig G, Weber A, Hassell N, Andrews J. Prevention of elbow injuries in youth baseball pitchers. Curr Sports Med Rep 8: 250–254, 2009.
7. Fleisig GS, Andrews JR, Cutter GR, Weber A, Loftice J, McMichael C, Hassell N, Lyman S. Risk of serious injury for young baseball pitchers. Am J Sports Med 39: 253–257, 2011.
8. Gandhi J, ElAttrache NS, Kaufman KR, Hurd WJ. Voluntary activation deficits of the infraspinatus present as a consequence of pitching-induced fatigue. J Shoulder Elbow Surg 21: 625–630, 2011.
9. Lyman S, Fleisig G, Waterbor J, Funkhouser E, Pulley L, Andrews J, Osinski E, Roseman J. Longitudinal study of elbow and shoulder pain in youth baseball pitchers. Med Sci Sports Exerc 33: 1803–1810, 2001.
10. Lyman S, Fleisig GS, Andrews JR, Osinski ED. Effect of pitch type, pitch count, and pitching mechanics on risk of elbow and shoulder pain in youth baseball pitchers. Am J Sports Med 30: 463–468, 2002.
11. Mullaney MJ, McHugh MP, Donofrio TM, Nicholas SJ. Upper and lower extremity muscle fatigue after a baseball pitching performance. Am J Sports Med 33: 108, 2005.
12. Murray TA, Cook TD, Werner SL, Schlegel TF, Hawkins RJ. The effects of extended play on professional baseball
pitchers. Am J Sports Med 29: 137, 2001.
13. Olsen S, Fleisig G, Dun S, Loftice J, Andrews J. Risk factors for shoulder and elbow injuries in adolescent baseball pitchers. Am J Sports Med 34: 905–912, 2006.
14. Posner M, Cameron KL, Wolf JM, Belmont PJ Jr, Owens BD. Epidemiology of major league baseball injuries. Am J Sports Med 39: 1676–1680, 2011.
15. Verducci T. Year-After Effect could strike many young arms in '07. 2006. Available at: http://sportsillustrated.cnn.com/2006/writers/tom_verducci/11/28/pitchers/index.html
. Accessed November 11, 2010.