Creating a logistical regression model to find out the likelihood of winning a category based upon how your team does.
Given what we know, can we predict that any categories are more or less important to win?
Call:
glm(formula = Win ~ R + HR + RBI + SB + OBP + K + QS + ERA +
WHIP + SVHD, family = binomial, data = df.n1)
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -10.01144 3.54634 -2.823 0.00476 **
R 0.08264 0.05334 1.549 0.12131
HR 0.18975 0.10480 1.811 0.07019 .
RBI -0.01158 0.04150 -0.279 0.78019
SB -0.01802 0.08653 -0.208 0.83505
OBP 15.52123 7.38919 2.101 0.03568 *
K 0.02986 0.02197 1.359 0.17409
QS 0.54061 0.18321 2.951 0.00317 **
ERA -0.57358 0.32174 -1.783 0.07463 .
WHIP -1.02193 2.17116 -0.471 0.63787
SVHD 0.29186 0.16172 1.805 0.07111 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 213.69 on 155 degrees of freedom
Residual deviance: 132.86 on 145 degrees of freedom
AIC: 154.86
Number of Fisher Scoring iterations: 5
What is the average score of all of a teams opponents by category.