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) -4.362437 2.441508 -1.787 0.07397 .
R 0.006073 0.036249 0.168 0.86695
HR 0.191374 0.081437 2.350 0.01877 *
RBI -0.008626 0.033230 -0.260 0.79517
SB -0.022390 0.069608 -0.322 0.74771
OBP 17.018019 5.965183 2.853 0.00433 **
K 0.008578 0.016497 0.520 0.60307
QS 0.195369 0.132965 1.469 0.14174
ERA -0.372551 0.239074 -1.558 0.11916
WHIP -3.030862 1.582270 -1.916 0.05543 .
SVHD 0.288158 0.120683 2.388 0.01695 *
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 295.26 on 215 degrees of freedom
Residual deviance: 210.71 on 205 degrees of freedom
AIC: 232.71
Number of Fisher Scoring iterations: 5
Call:
glm(formula = Win ~ R.R + R.HR + R.RBI + R.SB + R.OBP + R.K +
R.QS + R.ERA + R.WHIP + R.SVHD, family = binomial, data = df.n1)
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -23.253 5.612 -4.144 0.0000342 ***
R.R 5.127 1.496 3.428 0.000608 ***
R.HR 4.945 1.478 3.346 0.000820 ***
R.RBI 3.553 1.325 2.681 0.007338 **
R.SB 3.489 1.218 2.865 0.004170 **
R.OBP 5.363 1.477 3.632 0.000281 ***
R.K 5.120 1.460 3.507 0.000454 ***
R.QS 3.777 1.299 2.907 0.003645 **
R.ERA 5.574 1.571 3.549 0.000387 ***
R.WHIP 4.448 1.361 3.268 0.001085 **
R.SVHD 4.432 1.322 3.353 0.000800 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 294.130 on 214 degrees of freedom
Residual deviance: 51.745 on 204 degrees of freedom
(1 observation deleted due to missingness)
AIC: 73.745
Number of Fisher Scoring iterations: 9
What is the average score of all of a teams opponents by category.
Given the average team each week, how many times does a team score enough to win. Average team is found taking each weeks average for each category.
Spider charts with the ranking of each category vs. rest of the league.