Friday, July 18, 2014
WOWY: plain-english v not
Some ten or twelve years ago, I introduced WOWY (with or without you). The beauty of it is in the simplicity of description. The other beauty is I can drill down at a player or pair or any level I want. I control the entire process.
At its heart, it's nothing more than a regression with multiple variables to consider. So, it was interesting to see this article by Max Weinstein, and I'm sure Max Marchi did something similar two or three years ago (Marchi had asked me for the historical results of my WOWY that I published in the Hardball Times Annual so he had something to compare against, and I think the results were extremely highly correlated; I don't remember where Max published it, be it Hardball Times or BPro).
So, the reason I am not a fan of the approach that Max and Max use is because it is very black box. They rely on best-fit, and you may end up getting an absurd result. Think of an extreme example like Cal Ripken. I've seen the exact same approach from Brian MacDonald as he used it for NHL plus/minus. They all follow the same process. But I could tell by the results from Brian's work (tremendous as it is) that it didn't pass the sniff test. Some variable was not being handled or not handled well enough. You can check the old blog for that.
Anyway, so what you end up in the end is a list of results. And that's it. No real flavor for it, no profile. And it may end up being that the more automated approach that Max and Max and Brian et al follow could even be more precise, it's just a bit harder to make sense of it when something looks wrong.
?
Recent comments
Older comments
Page 1 of 151 pages 1 2 3 > Last ›Complete Archive – By Category
Complete Archive – By Date
FORUM TOPICS
Jul 12 15:22 MarcelsApr 16 14:31 Pitch Count Estimators
Mar 12 16:30 Appendix to THE BOOK - THE GORY DETAILS
Jan 29 09:41 NFL Overtime Idea
Jan 22 14:48 Weighting Years for NFL Player Projections
Jan 21 09:18 positional runs in pythagenpat
Oct 20 15:57 DRS: FG vs. BB-Ref
Apr 12 09:43 What if baseball was like survivor? You are eliminated ...
Nov 24 09:57 Win Attribution to offense, pitching, and fielding at the game level (prototype method)
Jul 13 10:20 How to watch great past games without spoilers