The throwing technique of a baseball fastball pitch can be described as a coordinated sequence of body movements and muscular forces with the ultimate goal of arriving at the highest ball speed possible at ball release [1
]. It is believed that the interaction of body segments transfers energy in a sequential pattern from the ground up to move the upper extremity joints into the right position to finally result in a high ball speed [2
]. The lower extremities and the trunk are the main force generators during initiation of the throw [4
]. For an optimal use of the trunk in force generation, the lower extremities have to be a stable base for the initiation of the rotations of the trunk and upper extremities throughout key phases of the baseball pitching action [5
When the leading leg is extended it is braced to enhance the ability of the trunk to rotate both forward and in the axial direction simultaneously [6
]. This braced leading leg was found to be associated with a high ball speed in pitchers [9
]. A detailed analysis of lower limb mechanics was performed in a study by Milewski et al. [10
], although they did not investigate the relation between lower limb mechanics and ball speed. A commonly held view is that pitchers who flex their knee after the moment of stride foot contact (FC) are not throwing to their highest potential [11
]. This is supported by a study that reported pitchers who threw at a higher velocity had both a slower rate of knee flexion of the leading leg on landing and a higher rate of subsequent knee extension as compared with pitchers that throw at a lower velocity [12
]. In addition to the knee angle at FC, greater knee extension of the leading leg was observed at ball release (BR) in faster throwing pitchers compared to slower throwing pitchers (32 SD 9° vs. 48 SD 14°) [6
]. Also, Werner et al. [13
] reported this association between the knee angle of the leading leg at BR and ball speed (β = −0.11, SD −0.029, p
= 0.009). In the same study, however, a more flexed knee of the leading leg at FC was associated with a higher ball speed, which contrasts with most other studies.
Another lower extremity parameter of interest in baseball pitching is stride length, which partly is dependent of body height and the build of the pitcher [14
]. A pitcher with a larger stride length results in more forward displacement, which can result in a higher ball speed [15
]. Furthermore, a larger stride length provides a greater moment arm for the trunk to rotate forward over the “locked leg”. Montgomery & Knudson [17
] observed a positive linear relationship (r
= 0.73) between stride length and ball speed in professional pitchers. However, this study was based on a small sample size and, to our knowledge, there are no other studies that relate stride length to ball speed, but a study has been conducted that showed that ball velocity does not have to be affected with a smaller stride length [18
The studies described above indicate the importance of the lower extremities in achieving high ball speeds in baseball pitching [15
]. Quantifying the associations of several lower extremity parameters at several instants in the pitching cycle might give additional insight in how the kinematics of the lower extremities contribute to ball speed. However, the existing literature that associates lower extremity parameters to ball speed involve only adult pitchers [6
]. Exploring the association between lower extremity parameters and ball speed in youth baseball pitchers might provide additional information as this study population shows more variance in anthropometric characteristics (segment lengths, force capacities) as well as in throwing technique and ball speed. The current baseball literature that includes youth baseball pitchers as participants is only descriptive without any associations with ball speed [10
]. Therefore, the purpose of the present study was to determine whether stride length and knee angle of the leading leg at foot contact, at the instant of maximal external rotation (MER) of the shoulder, and at ball release are associated with ball speed in elite youth baseball pitchers. It was hypothesized that a larger stride length, a more flexed knee at FC, and a more extended knee at MER and BR have a positive association with ball speed in elite youth baseball pitchers.
The mean ball speed observed for all 5 pitches of all participants was 67 mph (SD 8, range 48–82). The mean stride length of the participants was 140.8 cm (SD 15.2, range 99.8–173.8) and on average 79.8% (SD 6.0, range 62.4–92.8) of their body height. Results of the multiple linear regression analysis did not show a significant association between stride length as percentage of body height and ball speed (Table 1
; Figure 3
The participants had a mean knee angle at FC of 40.3° (SD 14.6, range 10.9–94.6). Linear regression analysis with body height and mound as confounding variables showed that knee angle at FC was not significantly associated with ball speed (Table 1
; Figure 3
The mean knee angle at MER was 45.0° (SD 17.8, range 9.7–96.0). Body height and mound appeared to bias the association between knee angle at MER and ball speed. After adjusting for these confounders, knee angle at MER was significantly associated with ball speed (Table 1
; Figure 3
C). The negative regression coefficient found for knee angle at MER indicates that ball speed decreases as the knee angle increase, i.e., as the knee is more flexed. The value of the coefficient (−0.055) shows that youth baseball pitchers who have the knee of their leading leg 1/0.055°, or ~18°, more extended at the moment of MER throw 0.45 m/s (1 mph) faster.
The mean knee angle of the participants at BR was 40.5° (SD 19.0, range 3.9–94.7). The association between knee angle at BR and ball speed appeared to be biased by both body height and mound. After adjusting for these confounders, knee angle at BR was significantly associated with ball speed (Table 1
; Figure 3
D). Youth baseball pitchers who throw with a 1/0.051°, or ~19.5°, more extended knee at BR pitch 0.45 m/s (1 mph) faster.
The purpose of this study was to determine whether stride length and knee angle of the leading leg at FC, MER and BR were associated with ball speed in elite youth baseball pitchers. In support of our hypotheses knee extension at MER and BR appeared to be significantly and positively associated with higher ball speeds.
The increase in ball speed, which is associated with a more extended knee at MER and BR, is relatively small. To achieve an increase in ball speed of 0.45 m/s (1 mph) a more extended knee of 18–19.5° is required. However, although the effect seems to be small, it still appears statistically significant in the population of the present study, which consists of a homogenous group of youth elite baseball pitchers. The observed result can be an indication of the relevance of knee angle. Moreover, the observed small effect may still have a practical relevance, because at the top level of baseball, small details can make a large difference.
The average stride length in this study was 80% (SD 6) of body height; this is comparable to the stride length found in two other studies [23
]. However, some other studies found different results [25
] (Table 2
). The latter studies defined stride length as the distance between the centers of the ankle joints and corrected this value for body height, while in the present study the distance between the pitching plate and the center of the ankle joint from the leading leg was defined as the stride length. Since most pitchers place the foot of their trailing leg in front of the pitching plate, the definition of these two studies results in lower relative stride length values. This should be considered for the comparison with the value of this study. The linear regression analysis showed no significant association between ball speed and stride length as percentage of body height in youth baseball pitchers (β
= 0.022). This association can only be applied to stride lengths within the range of 62–93%, because this study did not measure any stride lengths outside this range. Montgomery & Knudson [17
] is the only study that also examined the association between stride length and ball speed and they demonstrated a significant association between the stride length and ball speed. Pitchers in that study had to throw with their normal stride, with under-stride and with over-stride, which was a similar range (75–100%, assuming a body height of 180 cm) as in the present study. However, this study is not comparable with the present study because the association was determined for each pitcher individually, while in the present study associations were determined at group level. It might even be beneficial to have a shorter stride length, since it does not seem to affect ball speed, but it does reduce physical exertion [18
]. Overall, more research is needed to understand whether there is an association between stride length and ball speed or not.
The knee angle of the leading leg starts with a mean flexion of 40.3° (SD 14.6) at FC and is followed by more flexion at the time of MER (45.0° SD 17.8). Subsequently, the knee extends towards BR (40.5° SD 19.0), which is consistent with previously published results [26
]. However, the knee angle at FC is smaller compared to other studies [23
] (Table 2
). There are two studies which measured the knee angle at MER in youth baseball pitchers [24
]. They found a value of 46° (SD 15) and 39° (SD 12.1), which is a result that is comparable to this study (45.0° SD 17.8). The knee angle at BR is within the range of the values found in the other studies [10
]. We found a significant negative association between the knee angle at MER and BR with ball speed. This is similar to the study of Werner et al. [13
], which concluded that a higher ball speed (1 mph) is found in pitchers with more knee extension (9°) in the later part of the pitch cycle [10
]. According to these and our results, youth baseball pitchers should throw with a more extended knee of the leading leg. However, it is important to notice that the present cross-sectional study and the cross-sectional study of Werner et al. [13
] only report associations between ball speed and knee angle, which do not support a causal relationship in which a more extended knee would actually lead to higher ball speeds. Therefore, practical implications based on these associations should be critically evaluated. In case of a potential causal relationship, it should be realized that knee extension is limited, which means that the gain in ball speed by more knee extension in the later part of the pitch cycle is limited. It should also be mentioned that at maximal extension, the knee is more vulnerable to injury [27
The observed association between ball speed and the knee angle of the leading leg might actually be a causal relationship when several mechanical theories are taken into consideration. The extending knee results in a braced leading leg. This results in a braking effect during the stance phase, which means that the leading leg stops moving in a forward direction and the proximal segments rotate over the leading leg. The pitcher should not flex his knee from the moment of FC because this will result in energy dissipation since a non-moving and locked hip (i.e., a fixated trochanter major in space) requires less muscle power of the knee extensors with an extended knee compared to a flexed knee because of the shorter moment arm. The trunk is the segment with the highest mass and is, therefore, potentially one of the greatest force generators in the kinetic chain [4
]. Also, in other sports such as javelin throwing, the braking effect of the lead leg is shown to be important because it allows the trunk and upper extremities to accelerate forward over the leading leg, aiding in the transfer of momentum up through the trunk and the throwing arm [28
In the present study population of youth baseball pitchers, a large range of body types was present due to the obvious effects of growth and maturation at these ages [29
]. The effect of maturation was, however, not within the scope of the present study. If similar measurements would be performed over time, future studies could focus on the effects of growth in relation to throwing velocity. In the present study, however, we do have to correct for the observed range of body types to arrive at the independent association between the kinematic variables and ball speed. As explained in the methods section, we only used body height as a confounder, as it was expected to largely affect ball speed. Body height itself was highly correlated with body weight, age, and strength (r
> 0.75) (which were thus all strong predictors of ball speed (r
> 0.8)). Taking all these confounders into account at the same time in the multiple regression analyses would have introduced issues of collinearity. Therefore, only one of those potential confounding factors needed to be selected for eliminating confounding. Of those factors, body height appeared to be the strongest predictor and was, therefore, the variable chosen for exploring confounding. One should bear in mind that body height should be considered a variable representing other variables such as body weight, age etc., and not only an explanatory variable by itself. In the regression models it was observed that pitchers throw around 0.45 m/s (1 mph) faster for every increase in stature of 0.02 m. Another variable that was explored for confounding, and interaction, was the mound. Pitchers who threw off a mound had a more extended knee at the moments of FC, MER, and BR. Therefore, the mound was included as confounding variable. However, interactions with mound appeared not to be significant. This means that the associations between knee angle and ball speed, and between stride length and ball speed, are not different for the pitchers throwing from a mound and the pitchers not throwing from a mound. Others also reported kinematic differences between pitching off a mound compared to flat ground [31
]. Nissen et al. [31
] reported that the knee of the leading leg was in more extension at FC when pitching off the mound. This probably occurs as a consequence of the delay in lead foot contact when stepping down off the mound [32
]. Furthermore, ball speed has shown to be different when pitching from flat ground compared to pitching of a mound [33
]. This difference in ball speed shows the importance of including the mound as a confounding variable. However, having included pitchers that also do not throw from mound in the present study next to pitcher that do throw from a mound, warrants careful generalization of the results of the present study. In future studies, it should be considered to study the associations between kinematic variables and ball speed preferably when pitchers only throw from a mound, throw only on a flat ground, or do both.