Posted by tlowster on 4/22/2022 8:48:00 PM (view original):
I don't. I assume a neutral part and right handedness. Then I add points for lefties (about .007) and add for switch hitting (about .015). Also that formula above is more accurate for players near the 75/80 ratings. Once ratings get above or below that that, the formula under rates them. I could use a more statistics based formula with exponents, but I wanted it to be something that is quickly computed.

As far as stadium, I have mostly played in neutral parks to this point. I am experimenting in Colorado, but I'm only in season 1. I'm not too concerned with predicting batting average in extreme parks because I know they'll simply do better than my formula. I'm not doing the Colorado experiment to calculate staduim effects on batting average, though. Im doing it to test a pitching theory of mine. I'll likely need multiple seasons to test it to get rid of small sample size so maybe I can look at staduim effects to batting average as well soon.

I play in extreme parks and use a very similar formula to predict batting average to tlowster, I will say that it still does a pretty good job of predicting performance, although a minor adjustment is needed. I haven't done an actual statistical analysis to figure out how much of an adjustment should be made, but anecdotally I'd say it's about 20 batting average points. In other words, I've noticed that if the formula predicts a player to have a .250 batting average in a neutral park they would hit about .270 in an extreme hitter's park and about .230 in an extreme pitcher's park