L C Loram1, W McKinon, S Wormgoor, G G Rogers, I Nowak, L M Harden. 1. Applied Physiology Laboratory, School of Physiology, Medical School, University of the Witwatersrand, Johannesburg, South Africa. loramlc@physiology.wits.ac.za
Abstract
AIM: Studies investigating determinants of ball release speed have examined the technique and anthropometry of fast bowlers with little work being done on muscular strength. The aim of our study was to determine whether knee biomechanics during bowling and strength of the shoulder and knee could predict ball release speed. METHODS: Twelve cricketers, aged 16.6+/-0.7) years, from schools in Johannesburg, South Africa, volunteered for the study. Subjects were fast-medium bowlers (mean ball release speed of 29.2+/-1.8 m.s(-1)) and had been bowling for at least 5 years. Three accurate deliveries were filmed on an outdoor cricket pitch, in the sagittal plane with a high-speed digital camera recording at 250 frames per second. The mean ball release speed, knee angle at ball release and knee angle at front foot strike were determined using simple two-dimensional kinematics. On a separate day, peak concentric isokinetic muscle torque was measured for both knees and the dominant shoulder. RESULTS: Ball release speed was positively correlated to a straight knee at front foot strike (r=0.72, P=0.009) and at ball release (r=0.71, P=0.011). No significant correlation was found between ball release speed and any of the peak torque values (knee extension peak torque, r=-0.11, knee flexion peak torque, r=-0.08, shoulder internal rotation peak torque, r=0.21 and shoulder external rotation, r=0.29, P>0.05). A multiple regression model using knee angle at front foot strike and at ball release, and the angle at which peak torque is generated during shoulder internal and external rotation, predicted ball release speed (adjusted r2=0.85, P<0.002). CONCLUSIONS: We have confirmed that the angle of the front knee at the beginning and end of a delivery is an important correlate of ball release speed in schoolboy fast-medium bowlers. In addition we have also demonstrated that a multiple regression model based on knee kinematics and shoulder peak torque angles can be used to predict ball release speed.
AIM: Studies investigating determinants of ball release speed have examined the technique and anthropometry of fast bowlers with little work being done on muscular strength. The aim of our study was to determine whether knee biomechanics during bowling and strength of the shoulder and knee could predict ball release speed. METHODS: Twelve cricketers, aged 16.6+/-0.7) years, from schools in Johannesburg, South Africa, volunteered for the study. Subjects were fast-medium bowlers (mean ball release speed of 29.2+/-1.8 m.s(-1)) and had been bowling for at least 5 years. Three accurate deliveries were filmed on an outdoor cricket pitch, in the sagittal plane with a high-speed digital camera recording at 250 frames per second. The mean ball release speed, knee angle at ball release and knee angle at front foot strike were determined using simple two-dimensional kinematics. On a separate day, peak concentric isokinetic muscle torque was measured for both knees and the dominant shoulder. RESULTS: Ball release speed was positively correlated to a straight knee at front foot strike (r=0.72, P=0.009) and at ball release (r=0.71, P=0.011). No significant correlation was found between ball release speed and any of the peak torque values (knee extension peak torque, r=-0.11, knee flexion peak torque, r=-0.08, shoulder internal rotation peak torque, r=0.21 and shoulder external rotation, r=0.29, P>0.05). A multiple regression model using knee angle at front foot strike and at ball release, and the angle at which peak torque is generated during shoulder internal and external rotation, predicted ball release speed (adjusted r2=0.85, P<0.002). CONCLUSIONS: We have confirmed that the angle of the front knee at the beginning and end of a delivery is an important correlate of ball release speed in schoolboy fast-medium bowlers. In addition we have also demonstrated that a multiple regression model based on knee kinematics and shoulder peak torque angles can be used to predict ball release speed.