Wednesday, February 10, 2010

2010 NL Central - by WAR

By now, after 3 posts that professed my love of the stat WAR (Wins Above Replacement), it's probably evident that I am a supporter of it as a predictive tool.

Check the sidebar labels under "WAR" to see the 3 part series in which I explained WAR, examined how it did as a predictive tool in 2009, and how it may forecast the Pirates in 2010.

I decided to expand on that thought and use WAR to forecast all the teams in the NL Central. Now keep in mind, these are my estimates (not Fangraphs) based on past WAR totals for each individual player projected to start for the teams. After I did a detailed one for the Pirates, with each projected 25 man roster spot, a couple of things were evident:
1. Your bench does not contribute much WAR, due to lack of playing time. For the Pirates, the bench was 0.0. For the rest of the teams, I assumed a WAR of 1 for the bench, except the Astros got a 0.0
2. The bullpen does not contribute much either. Again, due to playing time (and the chaining effect detailed by Dave Cameron at Fangraphs), the bullpens ranged from 3-4 for the NL Central.

I did go player by player in a nerdy Excel spreadsheet for the rest of the starters, both batters and pitchers. Here are the overall results:
1. STL (48 baseline wins + 28.5 bat + 18 pitch) = 94.5
2. CHC (48 + 21.5 bat + 17 pitch) = 86.5
3. CIN (48 + 23 bat + 13.5 pitch) = 84.5
4. MIL (48 + 22.5 bat + 13.5 pitch) = 84
5. PIT (48 + 17.5 bat + 12 pitch) = 77.5
6. HOU (48 + 17 bat + 12 pitch) = 77

Some of the totals seem a little high to me, but remember that a +/- of 4 wins is very acceptable and easy to do over the course of a year. Also, part 2 of the WAR series showed that WAR was a great predictor of the "order" of the standings within a division, even if the totals were a little off.

When you do an exercise like this and expand your scope, you come to realize just how much more talent the Pirates need. That's no surprise in and of itself, but once you see in black and white how they compare, it's startling. Here is a position by position ranking of where the Pirates stand within the division.
C - 3rd of 6 (Doumit with 2, leader Molina STL 3.5)
1B - 6th of 6 (Clement with 0.5, leader Pujols STL 8)
2B - 2nd(t) of 6 (Iwamura with 2, leader Phillips CIN 3.5)
SS - 5th of 6 (Cedeno with 1, leader Ryan STL/Theriot CHC 3)
3B - 1st(t) of 6 (Laroche/Alvarez with 3.5, tied with Rolen CIN 3.5)
LF - 4th(t) of 6 (Milledge with 2, leader Holliday STL 5.5, but 4th/5th/6th all 2.0)
CF - 1st of 6 (McCutchen with 4, next closest Bourn HOU/Rasmus STL/Byrd CHC 3.0)
RF - 2nd(t) of 6 (Jones with 2.5, leader Pence HOU 3.5)

Without going into each SP, let me just put this into context a little bit. The Pirates top pitcher, Maholm with 3 WAR, would be tied for 8th. Carpenter and Wainwright for STL are both 5 WAR pitchers.

Although it doesn't seem like much, 1 and 2 WAR here and there add up over the course of a roster. As you can see above, the Pirates are woefully behind at 1B. The next closest to them is Berkman HOU and Lee CHC at 4 WAR. The Pirates need to inject talent wherever they can get it, but right now 1B, SS, and SP are the obvious areas.

That's why if Laroche improves his hitting and maintains his good fielding, he can be a 3-4 WAR player at 3B, allowing us to move Alvarez and his 4-5 WAR potential to 1B. If you add a competent SS at 2-3 WAR and a SP at 4-5 WAR, you have just added 7 wins to the roster. That's how the Pirates will break this losing skien that they have been on the past eternity.

If you have any questions about WAR's for specific players, let me know in the comments.

No comments:

Post a Comment