View Single Post
02-16-2014, 05:41 PM
Callahan Auto
Rational Police
Callahan Auto's Avatar
Join Date: Jul 2009
Country: United States
Posts: 1,601
vCash: 500
I don't mind oWAR, depending on the methodology. dWAR, though? I'd much rather use an eye test to assess defense. When using statistics to analyze defense, there's just too many assumptions being made to generate the output, in my opinion. It's part of the reason why dWAR is such a volatile stat on a year-to-year basis. Even Sabre guys that like dWAR prefer to use multi-year sample sizes because of the volatility. To me, that shows a weakness in the computations ability to interpret real-time performance.

Offensive numbers are controlled to the point where you can straight up compare players. Until every single ball in play is tracked and graded, defensive metrics will be too much guesswork to be taken for fact.

Callahan Auto is online now