The purpose of this study is to identify the risk factors which affect the consecutive onset of throwing shoulder injury of a collegiate baseball team, and to calculate regression formula which could predict the future onset.Prospective study. The subjects are 69 asymptomatic collegiate baseball players. We executed medical checkups and investigated who got throwing shoulder injury during this year. We did logistic regression analysis with the medical checkup findings and the data of the onset and found out the factors which significantly affected the onset. We also calculated Odds ratio of these factors and regression formula which predict the probability of the onset of throwing shoulder injury and evaluated the validity of regression model.52%(36/69 cases) of the players suffered throwing shoulder injury during this year. Among the medical checkup findings of asymptomatic phase, pitcher and catcher, past history of throwing shoulder injury, abnormality of scapula-humeral rhythm, heel-buttock distance were thought to be the risk factors which were near-related to the onset of throwing shoulder injury. We calculated regression formula using the medical checkup findings of asymptomatic phase and we could predict the onset of throwing shoulder injury with 82.5% accuracy.In order to prevent this injury, the method used in this study could be useful for field players to predict the onset of throwing shoulder injury.