Thursday, June 4, 2015

What is the best way to structure a baseball batting order?

I posted a slightly different version of this on

There frequently is discussion among baseball fans about the proper construction of a baseball team's lineup,

In the 1966 book, Percentage Baseball, by Earnshaw Cook (many copies available at, he shows that a lineup constructed from the best hitter to the worst will create, on average, at least 11 more runs a season. In 1956, Pirates Manager Bobby Bragan employed a lineup for the last 40 games of the season that followed that idea. It actually was Bragan's idea first. The Pirates won two more games in the last 40 than they had in the previous 40. But they were not a very good team, and Bragan abandoned the idea when he went to the Braves. He didn't think Aaron should bat first because he thought it denied him too many RBI opportunities. Cook shows that "if Aaron be moved from the lead-off position toward the fifth slot in the batting order, it will cost him from 3 to 14 runs contributed per season. How much it reduces team scoring will depend upon the relative performances of the other players."

Cook shows through probabilities that the best batting order has the team's best hitter hitting first, followed by the second best hitter, etc. He points out that each batter after the first, on average, has 17 few at bats during the season that the hitter immediately preceding him. And since the object is to score as many runs as possible, the best hitters should get the most at-bats. His probabilities show that this will result in more runs scored - and as the better the hitters on a team are the higher will be the rates of scoring.

"The interactions of baseball are entirely too complex to be appraised by the intuitive methods of any manager. If he wishes to gamble, as he must on every play, he might as well employ the most favorable odds. This is the only thing anyone can know (italics his) with assurance about baseball."

He goes on: "Winning baseball is associated with the favorable occurrence of very small differences of significant probabilities...."

Cook may have been the first to argue that the run differential is the most important statistic in baseball and that success of a team depends on maximizing that differential.

"The difference between a first-place ball club and the average four- or fifth-place team is entirely a matter of balance between the runs scored in 154 games by the team (R.t) and by its opponents (R.o)..."

Cook was probably the first to argue against the use of the sacrifice bunt. However, he was a big believer in the hit-and-run.

He was controversial in his time in the late 50s and early 60s but Frank Deford of Sports Illustrated wrote "Right now...Earnshaw Cook knows more about baseball than anyone else in the world....Nevertheless, baseball officials hesitate to consider his findings, and for a very good reason: if he is right, they have been playing the game all wrong for years."