Wednesday, December 30, 2009
Sabermetric Lesson: WAR
Alex is back for another lesson, this time, it’s WAR.
Buy The Book from Amazon
Alex is back for another lesson, this time, it’s WAR.
I have seen many more people use rWAR for career evaluations and more use fWAR for projections or to evaluate short-career pitchers, which seems appropriate. I agree that many people do not understand the differences between the different WAR models, and that we should be careful to use the best models for the best situations and discuss the differences to newbies.
Feb 08 15:55
Batman, the webslinger?
Feb 08 15:23
When is a life entity considered a person?
Feb 08 15:14
New PECOTA
Feb 08 14:44
When to purposefully lose the lead
Feb 08 13:49
The will of the people?
Feb 08 11:43
Is Nate Silver alot more certain than he lets on?
Feb 08 09:02
Forecaster’s Challenge: 2012?
Feb 08 07:43
For Your Soul
Feb 08 01:22
Why I’d Bet on My Model (and Against My Instincts)
Feb 07 20:05
Golfers “playing through”
I think it’s important to remind people that there are many interpretations of WAR. The construction of WAR basically goes like this:
Offense+Defense+Position+Replacement
You can include park adjustments, baserunning, league adjustments, quality of batters faced adjustments, etc. as well.
Offense can use anything you want, including WPA, wRAA, or even something like Runs+RBI’s above average. Defense can use UZR, Total Zone, FSR, +/-. Replacement level and positional adjustments can be anything you want, as long as you have some reason to back it up.
The point of this is that it bothers me to no end that people refer to FanGraphs WAR as “WAR”. That leads to people discrediting the stat itself because they don’t like FIP, or they believe context should be included.