Lineup optimization at the WCWS

Lineup optimization at the WCWS

In sabermetrics circles the debate on lineup construction is over. Using statistical methods such as Markov Chains, it’s possible to optimize a lineup. That doesn’t mean that all Major League Baseball teams utilize the advice of their analytics departments, since many old-school managers still exist in baseball and prefer the traditional method of filling out their lineup card.

After analyzing the eight teams in this year’s Women’s College World Series, it appears that some schools have done their research into sabermetrics.

Here is a comparison of traditional lineup construction versus the method from The Book that most sabermetricians favor:

Lineup Position

Traditional Lineup

Sabermetrics Lineup

1

This hitter has speed and  steals bases.

On-base percentage is key.

2

Good bat handler who can bunt.

One of the two best hitters on the team.

3

Hits for a high batting average.

Next best hitter after filling the #5 spot.

4

Hits for power.

One of the two best hitters on the team, but with more power than the #2 hitter.

5 Second best power hitter.

Next best hitter after filling the #2 and #4 spots.

6 6th best hitter.

Best remaining hitter by wOBA.

7

7th best hitter.

Best remaining hitter by wOBA.

8 2nd worst hitter.

Best remaining hitter by wOBA.

9 Worst hitter.

Best remaining hitter by wOBA.

Of the eight teams in the WCWS, Michigan comes closest to having a sabermetrics-based approach. With Sierra Lawrence (.553 OBP) in the leadoff spot and Sierra Romero (a team-leading .557 wOBA) in the #2 spot, Michigan’s offense is set up to be the juggernaut that they have been in 2016. From there Michigan’s lineup this week could look like:

Lineup Spot

Hitter

wOBA

3

Kelly Christner

.370

4

Kelsey Susalla

.428

5

Tera Blanco

.493

6

Aidan Falk

.338

7

L. Montemarano

.450

8

Faith Canfield

.290

9

Abby Ramirez

.376

You can see that with a few exceptions, Michigan is more like the sabermetrics lineup than the traditional lineup. By using a lineup simulator, one can see that optimizing Michigan’s offense would increase their overall run scoring by only a fraction over the course of the season by making a few minor changes. In other words the Wolverines are at the high end of lineup optimization in women’s softball.

The next team that comes close to optimizing their lineup is UCLA. The Bruins hit their best hitter, Mysha Sataraka (.507 wOBA), in the #4 spot and have Allexis Bennett (.500 OBP) as their leadoff so they’re off to a good start. The only change from a sabermetrics perspective would be switching their #2 hitter Kylee Perez (.375 wOBA) with their #3 hitter Delaney Spaulding (.437 wOBA). This change would have resulted in only about one more run for the Bruins this year, so not much of a change.

Georgia, like UCLA, bats their best hitter in the #4 spot. But Georgia hits their seventh-best hitter by wOBA in the #2 spot, meaning they’re treating #2 hitter Cortni Emanuel and her 32 stolen bases as a second leadoff hitter rather than as a spot for one of their two best hitters. My choice would be moving All-American Alex Hugo (.413 wOBA) to the #2 spot.. Nevertheless, at #15 for runs per game in the country this season, the Bulldogs are doing plenty of things right.

Two teams (Oklahoma, Alabama) bat their best hitter by wOBA in the leadoff spot and their second best hitter in the #3 spot. Meanwhile three teams (Auburn, FSU, LSU) bat their best hitter in the #3 spot.

It has been said that managers spend more time on their lineups for something that returns so little bang for the buck, and yet still often get it wrong. With that being said, I have taken enough of your time on the subject. Let the WCWS begin!