Ideas From the Fringe – Lineup Optimization

Oftentimes, things are divided between “old school” and “new school” when it comes to baseball strategies.  This is one of those times.

The Pirates are managed by Clint Hurdle who is a folksy-speaking, cliche-spewing, dirt-in-the-uniform guy.  He has control of the clubhouse and, along with his staff, have the Pittsburgh Pirates playing winning baseball deep into the season.

The Pirates have a general manager in Neal Huntington that is also a lifelong baseball guy, but he embraces the statistical, as well as the scouting side of baseball.  His love of numbers and stats is so great that it spawned this stat bouillabaisse of a quote at his introductory press conference in November 2007:

We are going to utilize several objective measures of player performance to evaluate and develop players. We’ll rely on the more traditional objective evaluations: OPS (on base percentage plus slugging percentage), WHIP (walks and hits per inning pitched), Runs Created, ERC (Component ERA), GB/FB (ground ball to fly ball ratio), K/9 (strikeouts per nine innings), K/BB (strikeouts to walks ratio), BB%, etc., but we’ll also look to rely on some of the more recent variations: VORP (value over replacement player), Relative Performance, EqAve (equivalent average), EqOBP (equivalent on base percentage), EqSLG (equivalent slugging percentage), BIP% (balls put into play percentage), wOBA (weighted on base average), Range Factor, PMR (probabilistic model of range) and Zone Rating.

Additionally, the Pirates have Dan Fox (previously a stat-heavy writer at Baseball Prospectus and Hardball Times) in the front office crunching data and producing models at a feverish pace.

The lineup that the Pirates put on the scorecard every night is a very traditional, old school-style of lineup.  I like to picture Dan Fox eagerly and hopefully checking the lineup when it is announced each afternoon, only to be crushed to see that it is one straight out of the 1970’s/80’s era of play.  He would be like a stat-nerd version of Tantalus, knowing he could help the lineup, only to continue to remain parched.

The authors of The Book has espoused lineup optimization for a few years.  It’s an idea that most of you are pretty familiar with and is pretty intuitive — you put your best hitters, regardless of body shape or power profile, high in the lineup to take advantage of their skills.  Additionally, the lineup’s ranking should be 1-2-4-5-3-6-7-8, in terms of placing the hitters by their ranked order.  Their research was based on thousands of simulations and research of years of data to determine scenarios that players encounter through the course of a game at various spots in the lineup.  Most versions of this focus on on-base percentage, but I thought I would take a stab at it using Weighted On-Base Average (wOBA) and breaking it down by lefty/righty pitchers.  The number in parentheses is the players wOBA split:

Versus RHP

1. A. McCutchen CF (.387 wOBA)
2. G. Jones 1B (.386 wOBA)
3. Walker 2B (.359 wOBA)
4. P. Alvarez 3B (.364 wOBA)
5. M. McKenry C (.342 wOBA)
6. T. Snider RF (.315 wOBA)
7. J. Mercer SS (.295 wOBA)
8. A. Presley LF (.282 wOBA)

Versus LHP

1. A. McCutchen CF (.494 wOBA)
2. S. Marte LF (.424 wOBA)
3. T. Snider RF (.379 wOBA)
4. M. McKenry C (.394 wOBA)
5. G. Sanchez 1B (.315 wOBA)
6. C. Barmes SS (.312 wOBA)
7. P. Alvarez 3B (.296 wOBA)
8. N. Walker 2B (.269 wOBA)

I couldn’t fully commit to putting the 4th best hitter in the 5th spot.  Just couldn’t go full-nerd on that one.  There are some sample size issues to consider here, as Marte/Snider/Mercer don’t have many at-bats to work with.  Also, there are only 3 players that have more than 100 at-bats against a LHP this season.

Take a gander at what McCutchen has done against LHP this year.  Keep in mind that wOBA has a very similar relevance scale to OBP — average is .330 to .340, .350 to .360 is pretty good, and so forth.  To see McCutchen with a .494 wOBA is stunning.

Having McKenry as the cleanup hitter against LHP is not conventional in any sense, but it’s more a function of how anemic the remainder of the lineup is against lefties.

The lineup versus RHP is not that ridiculous of a lineup in terms of old-school sensibilities.  You have speed at the top with McCutchen, a good overall hitter in the 3 spot with Walker, power with Alvarez in cleanup and RBI-collectors at 5 and 6 with McKenry and Snider.  If you catch Hurdle at just the right moment when one of the moons of Jupiter passes in front of Mars, you might be able to convince him to run that lineup out one night.

But don’t hold your breath.  Although it would be nice to reward Dan Fox’s faith in numbers just one night at least.


About Kevin Creagh

Kevin Creagh resides in the suburbs of Pittsburgh and has been a Pirate fan since the mid 1980′s. Kevin joined Pirates Prospects in July 2010 and enjoys writing about the economic side of baseball decisions. Many of Kevin’s articles focus on the long-term decisions that the Pirates and other organizations may need to make in the future, in order to balance escalating player salaries with built-in payroll limitations. Kevin also enjoys following prospects throughout the Pirates’ minor league system and forecasting their potential to help the major league team.
  • John Lease

    There is no school, old, new or middle, that advocates playing the twin towers of turd, Barajas and Barmes, as much as they do. Barmes hit 2nd yesterday, and actually got a hit AND walked twice. An offensive explosion for him, certainly. Nevertheless, his long body of work is who he is, a crappy hitter with a teensy amount of power. And he’s light years ahead of Barajas, who can’t even play defense anymore. For my sanity’s sake, no more than a game or two a week for them.

  • leadoff

    Stats have been around since baseball was invented, common sense has been around since baseball was invented, in my book you need both.

  • Richard Ya’Zhynka

    I’d like to see the Pirates use a lineup created by Dan Fox. The authors of the book would also have the pitcher batting eighth.

  • Brian

    I would like Snider batting leadoff at this point if he could stay in the lineup. Only Snider and Alvarez have over a 10% walk rate. Before this season I would have liked McCutchen at leadoff but #2 is more appropriate with his increased power.

  • e poc

    I seriously doubt Dan Fox would advocate making decisions based on tiny samples of platoon splits. A full season isn’t enough of a data point from which to infer true talent, and in most of these cases (McKenry, Marte, Snider, Mercer, Sanchez, Presley) you’re dealing with samples much, much smaller than that. It’s cool that you’re thinking outside the box a bit, but this isn’t the right way to do this analysis.