Looking at the lefty/righty splits for the Nationals outfielders

The Nationals’ crowded outfield includes Jayson Werth, Mike Morse, Roger Bernadina, Rick Ankiel, and Nyjer Morgan.

Using a naïve approach we may rank the outfielders by WAR in which case we would say that the starting three should be Werth, Morse, and either Ankiel or Bernadina. The same conclusion would be drawn by simply analyzing each player by his wOBA.

But several Nationals outfielders face extreme offensive splits lending support to a platoon. The career wOBA splits of the Nationals outfielders look like this:

Versus L:

1. Werth (.401)

2. Morse (.385)

3. Bernadina (.328)

4. Ankiel (.301)

5. Morgan (.261)

Versus R:

1. Werth (.353)

2. Ankiel (.341)

3. Morse (.336)

4. Morgan (.333)

5. Bernadina (.293)

Meaning that if the outfield was decided purely on offense it would look like this:

vs L:

vs R:

But even sabermetricians now realize that offense is not everything. Morgan was a fantastic fielder in 2009 and Morse doesn’t seem to be able to field anything. I used each player’s career UZR/150 in the corner slots and in center to complete some very rough calculations of WAR (See war_approximations.xls) to help see who would be the best choice in each spot after factoring offense and defense. This is what I got:

vs L

vs R

And then when I thought I was finished my friend raised a good point: Morse could platoon with LaRoche—who struggles against lefties—at first. Adding this scenario to my analysis I came up with the following arrangement:

vs L
vs R

While all the arrangements are worth commenting on, I’m going to restrict myself to the last one as it is theoretically the most complete.

  • ·Morgan’s defense is superior in the corner (25 UZR/150 vs 18 in center) giving him a large defensive advantage in there. When the positive positional adjustment for playing center and the negative adjustment for the corner are factored in, Morgan’s value becomes about equal regardless of where he’s playing.
  • ·Although Werth does play better defense in the corner, his UZR/150s are not as extreme (11 in the corner vs 6.1 in center) meaning the difference in his fielding values is also not as extreme as someone like Morgan. This means that after the positional adjustments Werth becomes more valuable in center.
  • ·Morse will likely play worse defense than LaRoche but he has such a strong offensive advantage over Old Smokey against lefties (a .385 wOBA for Morse and a .326 wOBA for LaRoche) that he becomes the better play with a southpaw in the mound. This opens up a corner spot for Bernadina.
  • ·Ankiel’s relatively strong wOBA against righties (.341) ensures him a spot with righties on the mound.
  • ·I appreciate Nyjer Morgan a lot more after doing this. His defense is very good. I’m also finished with questioning the Ankiel signing. It seems like a smart gamble.
  • ·Most importantly, this makes me rethink how lineups are determined. Just because we think of Nyjer Morgan as a center fielder may not mean he’s most valuable to his team playing that position. Fielders should be considered in terms of all the positions they can play and in comparison with all other players on the team who can field those positions. This can lead to some counter-intuitive results—I wasn’t expecting Morgan to end up in the corner.

Even if my findings were highly accurate (which I wouldn’t count on; if you look at the data, you’ll see that my WAR values make no sense except for doing rankings) there are two main reasons why neither of these lineups would be used in practice. The primary, and less tenable, reason is popular opinion would not support them. Imagine the mass criticism Jim Riggleman or Mike Rizzo would receive if LaRoche were benched against lefties or Werth were played in center over Morgan.

The second and more readily defensible reason is uncertainty. We do not really know how well Morse will play first or whether Werth will be able to handle playing center on a regular basis.

The risk, however, does not actually seem that big with Werth being the largest question mark. (I would think that it would become rapidly clear whether he could handle center or not and the team could move from there.) Yet the difference between each of my proposed lineups and the next best alternatives (which tend to be more realistic) are so small that it may not be worth it for the Nationals front office to undergo the criticism that would come with such an arrangement. But I am a firm believer that a team should do whatever it takes to win even if it gets them laughed at. And if you’re going to get laughed at either way, why not try it?