Carlos Santana played a good game offensively in the Cleveland loss on Sunday. He singled, homered and drew a walk, leaving him with a .231/.364/.432 slash line. That reminded me of a concept >Bill James introduced in his early Abstracts, batting average as a percentage of offense. At that time (the early 1980s), fans were more likely to look at a player’s BA than his OBP or slugging percentage. By looking at the percentage of offense due to batting average, you could get an idea of how much the player contributed beyond singles.

James did this graphically with rectangles. The base of the rectangle represented the OBP of the player, the height the slugging percentage. The area of the rectangle represented the full value of the player. Inside that rectangle another was drawn, with OBP only using hits in the numerator, and batting average as the height (removing isolated power component of slugging percentage). The area of that rectangle represented the value of a player’s batting average, or more precisely, his hits. Dividing the BA Area by the O*S Area, we get the percentage of offense due to batting average.

As you can see in Carlos’s slash line, his percent of offense due to BA is very low. The following table shows the 15 lowest percentages this season:

Note that his says nothing about the quality of the player. Jose Bautista is one of the best hitters in the game today, Adam Dunn’s season leaves much to be desired. With the MLB average of OBP at .319 and slugging at .391, an average OS Area would be .125. Thirteen of the players in this list are above that level, suggesting they indeed are hitting well. Santana is such a player. Just looking at his batting average indicates he’s not all that good. All the walks and extra base hits are very valuable to the team. In the days before the majority of teams used advanced analysis to value players, Santana would be a cheap pickup for a smart team. They would look at his batting average as a weakness, without noticing the strengths in the other parts of his game.