Line Up Construction 101 – Some Theories, Some Research, Some Thoughts

managers book

As Many Options as There are Stars in the Sky

In Bill James Guide to Baseball Managers from 1870 to Today (Scribner, 1997), James examines the multitude of options a manager has when constructing a line up.  He writes on page 20 that there are 741,354,768,000 different ways to create a nine player line up from a twenty five man roster.  Then, taking into account factors such as defensive position the options are reduced.  “If you have a roster of fourteen position players and eleven pitchers and you assume that only one pitcher will start, that reduced the options for the starting lineup from seven hundred billion to one billion.  If you assume that only certain players can catch, only certain players can play the outfield, etc., that reduces the options further; heck you can get down to a few million in no time.”

A few million options for a team’s manager to decide from seems more theoretical exercise in thinking than true game planning but there are in fact millions of options that can be picked from.  Helping to reduce the demand on one’s brain is the reality that teams often have at least one or two superstars locked in at a certain fielding position (ex. Joey Votto playing first base for the Reds) and/or a starter that is clearly the number one option when compared to other pitching options (Madison Bumgarner of the Giants).  The most ‘play’ a manager appears to have is with the batting order and the value he places on batting versus fielding options or platoon possibilities when determining his optimal line up.

bumgarner  votto-2


The Impact on Statistics Based on Where One Bats in the Order

There is a good deal of research that has been completed in recent years looking at the effect that the spot in a batting line-up has on a hitter’s statistics.  The research is clear – batting earlier in the line-up leads to more at bats over the course of a season (gaining an estimated 14 to 18 at bats per slot with each move up the order – going from batting 9th to 8th to 7th etc.).  Moving from 9th to 1st for example would lead to an estimated additional 112 to 144 at bats.  That is an extremely significant difference that is not lost on intelligent front offices and managers.

To take advantage of these additional at bats, managers have to use the right assessment tools in order to select an appropriate candidate.  For instance, using On Base Percentage (OBP) has become more widely accepted as a measure for determining who might help the team the most with an additional 100-140 at bats in a season while batting average, seen by many as a ‘noisy’ statistic that is likely to fluctuate more than the OBP statistic as well as a statistic that misses a large chunk of the ability of a batter to get on base, is used much less often.  Still, there are some managers who pay less attention to OPB when selecting a lead off man than their peers (see Dusty Baker’s use of sub.300 OBP Ben Revere in 2016).  Batting a sub .300 OBP hitter first or second in the order, regardless of the other skills that this player might possess, is to the detriment of the team’s offensive output.


Additionally, raw speed is no longer enough for a player to grab the leadoff spot.  Now, managers need to consider stolen base opportunity and success rate as well as a player’s overall ability on the base paths (ex. the ability to successfully take an extra base – such as going from 1st to 3rd on a single with regularity).  You likely won’t see a leadoff hitter with stats like Omar Moreno’s 1980 season (.306 OBP, 96 stolen bases/33 caught stealing, and striking out approximately 2 times for each walk (101 vs. 57).  96 steals look nice on paper but succeeding less than 66% of the time a steal is attempted is detrimental to a team’s ability to score runs especially when that lead off man is only on base approximately 30 percent of the time in the first place.


Research also indicates that the place where one bats in the line-up significantly impacts multiple counting statistics including Runs scored and Runs Batted In.  There seems to be some debate regarding the number of stolen bases attempted based on where a player bats (T. Cockcroft doesn’t find line up spot significant in relation to the stolen base but T. Bell who uses a more in depth calculation formula finds a significant difference – in reviewing the methodology of both I tend to give more weight to T. Bell’s results).

Future Research Incorporating All of the Above

Given the above, the manager (or front office) has a good deal of influence on both player and team production when making decisions on where to bat each player.  An interesting study to examine line up construction’s impact in a pennant race would be to examine the 2007 Phillies and Mets.  The Phillies erased a 7 game deficit in the final 17 games and ultimately captured the N.L. East pennant.  Could a different line up construction for the Mets during the final 17 games have kept them atop the East?  Did the Phillies maximize their offense in just the right way as they made their final push?  It would be very interesting to explore what decisions were implemented and then evaluate that process rather than just relying on the final standings to determine whether rosters were used to maximum effectiveness or not.  Was the Mets’ line up construction sound and their failure based on atypical lack of production or did the Mets mismanage their line up and ultimately have only themselves to blame for their collapse?

Proactive and Reactive Decision Making

Second guessing line-up construction and in game decisions are two favorite activities for disgruntled fans and armchair G.M.s.  When looking at manager competency, perhaps it is best to separately examine proactive decision making abilities (line up construction) versus reactive decision making abilities (in game moves).  It seems likely that these are two very different managerial abilities and ballgames can be won or lost based on either or a combination of these two decision types.

With hundreds of thousands/millions of options available to consider, it is little wonder why each person believes their strategy makes them the smartest person in the room and why other guys/teams just don’t have a clue.  Research has shown the impact of both line-up construction and, to a lesser extent, in game decision making.  Although a good deal of line up construction research has been conducted in regards to player valuation within fantasy baseball, the results have implications that cannot be ignored by MLB teams.  A more systemic analysis of line up decisions could yield a tremendous amount of data and help better evaluate managers and front offices in the real world.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s