January 4, 2007

2006 Lineup Analysis

WHYGAVS did a very nice piece called Fun with lineup tools where he used the ZiPS projections and fed them into a lineup tool called Baseball Musing's Lineup Analysis tool. A very fine read, please check it out before you read further.

I decided to take it a step in the opposite direction using the lineup he started with, his second lineup and a lineups including Laroche using 2006 stats instead of ZiPS. I also used the same model he did. For Nady's stats I used his actual stats he accumulated with the Pirates .353 OBP and .409 SLG.

As a point of reference the Pirates scored 691 runs in 2006.

In the first scenario I used the same lineup that WHYGAVS used. Using the ZiPS projections he came up with a 4.361 runs per game average. Using 2006 stats the same lineup had a 4.301 runs per game average. This adds to 697 runs for the season.

In the second scenario he substituted Bautista for Castillo, so I followed suit. ZiPS projection came out to be 4.461 runs per game. Using the 2006 stats the same lineup had a 4.470 runs per game. That is 724 runs for the season.

In his third scenario he added LaRoche to the mix. The ZiPS lineup came in with a 4.644 runs per game. Using the 2006 stats the same lineup came in with a 4.800 runs per game average. That figure to 778 runs for the season.

In his forth scenario he dropped Castillo into Duffy's spot coming up with a ZiPS projection of 4.609. Using 2006 stats actually raised the runs per game to 4.823. This figures to be 781 runs for the season.

The numbers across the board are quite similar is the first and second set of stats. This outcome didn't surprise me much since the ZiPS projections were fairly flat for most of the players.

The surprise that came from my poking around was the third set of stats showing a .156 improvement. That is 26 runs more on the season than WHYGAVS findings.

Another surprise with the forth scenario where the runs per game actually rose even further by trading Duffy and installing Castillo.

One other lineup I ran was as follows: Bautista, Wilson, Sanchez, Bay, Nady, Doumit, Castillo, Paulino, Duke. Basically dropping Duffy and replacing him with Bautista. The runs per game jumped from 4.301 to 4.431. The season run total went from 697 to 718.

The only things these findings show to me is the overvaluing of Duffy that has happened throughout the fan base and the press. This statement is probably an exaggeration because Duffy's speed is not taken into account in the stats used. It also does not take into account his defensive prowess.

The other thing it shows is the true need to add a power bat. However since I did use the pitiful power stats Nady put up offensively with the Pirates, there is likely going to be some improvement from him and hopefully Doumit.

One other thing this shows me is the fact ZiPS doesn't change a teams stats all that much from season to season. With as many young players as what is on the Pirates, there is likely to be improvements from Bautista, Doumit and Nady. The players built for a dropoff are Sanchez and Paulino.

The true question marks are Castillo and Duffy. I think these two players really determine the fate of the offense in 2007 especially if no lineup changes are made. If we could only trade those two for LaRoche.
Post a Comment