The Beltran debate really is part of a much larger discussion on the subject of lineup optimization. The fundamental question is (or should be) "How might the manager sequence his hitters to score the most runs?"
The ground-breaking analysis in The Book: Playing the Percentages in Baseball by Tom Tango, Mitchel Lichtman and Andrew Dolphin notes that the difference in runs scored by various lineups is not as dramatic as one might expect. In fact, any edge gained or lost in tinkering is very small.
At Beyond the Boxscore, Sky Kalkman summarized The Book‘s view of the subject in a position by position breakdown. In a nutshell, more importance is assigned to the second, fourth and fifth spots, and less to the three hole. One of Tony La Russa's favorite slump-busters, the pitcher hitting eighth, is preferred.
Kalkman's summary: "So, you want your best three hitters to hit in the #1, #4, and #2 spots. Distribute them so OBP is higher in the order and SLG is lower. Then place your fourth and fifth best hitters, with the #5 spot usually seeing the better hitter, unless he's a high-homerun guy. Then place your four remaining hitters in decreasing order of overall hitting ability, with basestealers ahead of singles hitters."
David Pinto at Baseball Musings automated this logic, developing a lineup optimizer tool. Enter player OBP and SLG and out comes a list of most (and least) optimal lineups. Not included as input to the calculations is the handedness of the batter or his speed, which are limitations, in my opinion.
Just for fun, I entered into the tool the career* OBP and SLG of a projected 2012 Cardinals lineup favored by one reader passionate on this subject. (* For the pitcher's spot only, I used 2011 actuals.)
One could suggest that the older players may deliver less than their career averages this season while the younger players could improve. Certainly that is possible, but I wanted to avoid using just one season's worth of data as it could unfairly skew the outcome.
However, one could enter your favorite projections into the tool to develop the best lineup using those, or any, inputs.
Here are the players' career numbers. This lineup as ordered would generate 4.911 runs per game.
|9||* Pitcher 2011||0.181||0.199|
|4.911||Runs per game|
The optimizer named the following as the best possible lineup with the assumed nine players at 5.151 runs per game. In other words, this batting order would theoretically score one more run every fourth day compared to the one I entered above.
Further, in a reminder how thin this bologna is sliced, the difference between the best and 30th-best lineup is just .011 runs per game – one run every 91 contests.
|5.151||Runs per game|
This lineup is most unusual to say the least, especially Lance Berkman hitting first. In fact, the best 30 lineups all have him in that position in the batting order. Without exception, Matt Holliday and Carlos Beltran are either second or fourth. Other certainties across all of the top-scoring 30 lineups were the pitcher hitting eighth and Descalso ninth.
Stepping back a bit, the concept of ensuring as many at-bats as possible for one's best hitters by batting them higher in the order would seem consistent with common sense.
Of course in reality, the only man that matters, Matheny, won't be batting Berkman leadoff. Further, he is not yet showing his hand about where he will hit Beltran in his regular-season lineup.
Kalkman's closing was the best reminder of all to keep this all in proper perspective: "Finally, stop talking like the lineup is a make-or-break decision."
The batting order offers good fodder for discussion, but the flap made about it is almost always much greater than its actual importance warrants.
Brian Walton can be reached via email at firstname.lastname@example.org. Also catch his Cardinals commentary daily at The Cardinal Nation blog. Look for his weekly minor league content during the season at FOXSportsMidwest.com. Follow Brian on Twitter.
© 2012 The Cardinal Nation, thecardinalnation.com and stlcardinals.scout.com. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.