• Journal of the Korean Data and Information Science Society
• /
• v.25 no.3
• /
• pp.485-492
• /
• 2014

In baseball, the traditional measure of pitchers are wins and ERA. But these statistics are influenced by luck or team power. So sabermetrician proposes a number of indicators that predict future performance. We determine a new measure, which we call pitching grade index (PGI) that efficiently summarizes a pitcher's performance on a numerical scale using principal components analysis. The PGI statistic can often be useful to assessing a pitcher's individual contribution. Also K-means clustering algorithm are used for segmentation of players into groups.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.2
• /
• pp.357-363
• /
• 2014

OPS is a sabermetric baseball statistic calculated as the sum of a player's on base percentage (OBP) and slugging percentage (SLG). One of the frequently cited problem with OPS is that OPS gives equal weight to its two components, OBP and SLG. In fact, OBP contributes significantly more to scoring runs than SLG does. This paper provides some exploration into the correct weighting of OBP to SLG when adding the two together. By correlating different coefficients of OBP to runs scored per game, the weighted OPS that weighting OBP 56% in two place more than SLG produced the highest correlation. We found that the weight of OBP increases as RPG increases. Also we suggest the linear regression equation of the best OBP coefficient against RPG.