빅 데이터 분석 기법을 이용한 한국프로야구 주자 평가 모델 = Efficient estimation model of runner using big data analysis in Korean baseball league

다국어 초록 (Multilingual Abstract)

In this paper, we proposed a new type of Runner Estimation Model that takes into account the batting position, the left/right characteristics of the pitcher, and the victory contribution according to the game situation using the big data analysis technique in the KBO League. The proposed model consists of a Winning Contribution Estimation Module and an Advance Estimation Module for left/right characteristics of the pitcher.

The Winning Contribution Estimation Module consists of a Steal Victory Contribution Score and an Additional Victory Contribution Score. Steal Victory Contribution Score refers to the score that contributed to the victory of the team according to the situation before and after the runner's stolen. The Additional Victory Contribution Score is the score that contributes to the victory of the team due to the additional runs by comparing the pre/post game situation of the runner. And the Advance Estimation Module is a module that calculates the advance score by using the difference between the average ability of all runners and the ability of each runner to advance runs according to the batting position.

In order to verify the effectiveness of the proposed model, we compared the rankings of KBO League Records in Korea. As a result of the experiment, unlike the KBO League Records, which ranked in terms of the number of successful stolen bases, in this study, runners who have contributed significantly to team victory are in the top rank. In addition, the runners were ranked differently depending to the left/right characteristics of the pitcher. As a result, the proposed model is considered as a new Runner Estimation Model that can effectively evaluate the performances of the runners who have contributed greatly to the team win.

The proposed model can analyze the competence of the runners according to the items and apply them to the individual training or the strategy development process according to the left/right characteristics of the pitcher. It can also be used for salary negotiation through rank evaluation of all runners. Further research is needed to develop a model that takes into consideration the ability of the defenders to stop the offense, which affects the runner's ability to move forward.