Originally Posted by Cooper
I was going thru Bill James Online site this morning and realized that Chapman lead all NL pitchers with Win Shares at 21 and finished 2nd in the major league (Verlander finished first).
I have a limited working knowledge of Win Shares but i do believe they are a better indicator of value than WAR measurements. I'm gonna take a stab as to why Chapman's win share value is so high.
1. James believes that some situations are leveraged more than others.
2. The ballpark in which the player plays in gets taken into account better than the WAR system: something along the lines of Chapman saving a run in a high run environment is worth more than in a low run environment. Since GABP is a hitters park Chapman's production is worth more than say Kimball -who pitches in a park where runs are scarce. I know the WAR system takes that into account, but i really do believe they WS system does a better job of evaluating value.
Maybe the Reds are maximizing Chapman's value by keeping him in the bullpen.
One thing that bugs me about giving credit for leverage is that the situation in which a team chooses to use a player is not a characteristic of the player itself. That is to say, you can get more win value out of a player by using him in those situations, but that extra value is available to any player used in those situations. As a GM, leverage represents an increased opportunity/risk to get more win value from a given talent level.
Thus, I hesitate to give a guy extra credit simply because a team can choose to leverage him a certain way -- you could do the same for any player and it all cancels out in the end. Sure, you need to be thoughtful as a GM to ensure you'll have the opportunity go get the most out of your roster. For example:
- you shouldn't sign three 6-win 1B since you couldn't play them all and get full value
- you shouldn't spend money on another low OBP slugger if you don't have sufficient baserunners to drive in
But once you get past the obvious, most of the gains in this area are pretty marginal.