Wednesday, July 21, 2021
Behind the wOBA curtain
Patriot has been doing great work on all things Runs Created (and Above Average). In this episode, he brings in wOBA.
***
I'd also say the largest benefit (which I didn't think of at the time, but appreciated afterwards) of wOBA: having a "negative" RAA is still valuable, which wOBA "hides" because of its scale.
Imagine if instead of ERA we'd make RAA as the central stat for pitchers: you'd have someone with an ERA 5% worse than league average on 200 IP come out as "worse" than someone with an ERA 10% worse than league average on 50 IP.
So I think the main issue is that once you combine two dimensions (quality and quantity) into one dimension (RAA), you necessarily told reader you made a choice. Which is ultimately what slowed down the adoption of Pete Palmer's method.
RAR (or WAR), gave us that happy medium where we could combine things into one dimension, and so filled that gap of RAA which is too extreme one way for some and RC which is too extreme the other way for others.
0 = valueless is a useful construct