On the Quasisubjectivity of Pool Stats Data Collecting

PoolStats

Pool Stats LLC
Silver Member
XgbkJtVl.png


ABSTRACT:
There have been a few outcries on how we handle our data collection. Mainly, there is a notion that our data entry is subjective and will not generalize to a larger degree of players. This paper seeks to address these notions about how we handle subjective input and the policies, guidelines, and algorithms we have in place to ensure generalized and objective data. In order to accommodate a larger reading audience, we keep this paper brief on the math since the concepts are relatively easy to understand and further rigorous explanation is unnecessary.

LINK TO PAPER:
http://poolst.at/pdf3
 

Cron

AzB Silver Member
Silver Member
I read you're paper, but before anything I must say that if you receive feedback and won't change, then just state you don't care for feedback. The entire document seems to be in defense of 3.2.10, which doesn't need justification if that is what you're going to do anyways. No system is perfect, so in that way you're system is equal to anyone else's and you should be proud of that.

I won't name competing statistical systems, but there's a system around here that uses a "clutch" variable that seems compelling as it doesn't take into account defensive play, although ironically I'm not too sure if most would agree on what is the most clutch, thus enters opinion. Regardless, the rest of the system seems pretty good since its nature is to resist extremely subjective variables like "knowledgeable announcers". Even if announcers are always right 99% of the time, why allow that 1% in, where does it stop?

But if you'll never change, then state that. At least that way you can be a solid static reference which, in my opinion is a good thing. After all, people expect 12 inches in a foot to never change.
 
Last edited:
Top