Abstract
This paper offers some long term perspective on what has been happening to some baseball statistics for Korean professional baseball. The data used are league summaries by year over the period 1982-2013. For the baseball statistics, statistically significant positive correlations (p < 0.01) were found for doubles (2B), runs batted in (RBI), bases on balls (BB), strike outs (SO), grounded into double play (GIDP), hit by pitch (HBP), on base percentage (OBP), OPS, earned run average (ERA), wild pitches (WP) and walks plus hits divided by innings pitched (WHIP) increased with year. There was a statistically significant decreasing trend in the correlations for triples (3B), caught stealing (CS), errors (E), completed games (CG), shutouts (SHO) and balks (BK) with year (trend p < 0.01). The ARIMA model of Box-Jenkins is applied to find a model to forecast future baseball measures. Univariate time series results suggest that simple lag-1 models fit some baseball measures quite well. In conclusion, the single most important change in Korean professional baseball is the overall incidence of completed games (CG) downward. Also the decrease of strike outs (SO) is very remarkable.
본 연구에서는 한국프로야구 변천사를 야구 통계량들을 중심으로 살펴보았다. 분석방법으로는 1982년부터 2013년까지의 한국프로야구 데이터를 이용하여 야구 통계량들의 시계열 그래프와 상관계수를 이용하였다. 그 결과 유의수준 1%에서 연도와 유의한 양의 상관관계를 보인 통계량은 2루타, 타점, 4구, 삼진, 병살타, 사구, 출루율, OPS, 방어율, 폭투, WHIP이고, 유의한 음의 상관관계를 보인 통계량은 3루타, 도루자, 실책, 완투, 완봉, 보크였다. 상관계수가 유의한 야구통계량의 예측을 위해서는 Box-Jenkins의 ARIMA 모형을 이용하였다. 결론적으로 세월의 흐름과 가장 상관이 큰 것은 완투 횟수의 감소이며, 그 다음으로 삼진 개수의 증가를 들 수 있었다.