Packages Needed: tidyverse
library(tidyverse)
library(readxl)
WNBA <- read_xlsx("Data/WNBA.xlsx")
NBA <- read_xlsx("Data/NBA.xlsx")
WNBA400 <- WNBA %>%
filter(MP >= 400)
NBA1000 <- NBA %>%
filter(MP >= 1000)
## (Intercept) FG FGA FT FTA ORB
## -13.48442326 -0.14858126 -0.98827249 0.24640792 -0.44554616 0.46914799
## DRB AST STL BLK TOV PF
## 0.21856746 0.52798205 0.80738069 0.25279600 -1.60086566 0.04301587
## PTS
## 1.00000000
## (Intercept) FG FGA FT FTA ORB
## -6.5224804 -0.2960982 -0.8131592 0.1429953 -0.4431946 0.5238405
## DRB AST STL BLK TOV PF
## 0.2090862 0.5620121 0.8658215 0.3371603 -1.5332580 -0.1401101
## PTS
## 1.0000000
## (Intercept) FG FGA FT FTA ORB
## 0.4837048 1.9928367 0.8228087 0.5803193 0.9947221 1.1165783
## DRB AST STL BLK TOV PF
## 0.9566211 1.0644530 1.0723832 1.3337250 0.9577680 3.2571728
## PTS
## 1.0000000
## (Intercept) FG FGA FT FTA ORB
## 0.68 26.77 35.58 26.32 25.47 42.74
## DRB AST STL BLK TOV PF
## 11.45 29.70 47.30 38.00 47.90 18.80
## PTS
## 41.10
I took a regression of the stats included in PER against the wins score. I then took the coefficients and scaled them against the largest coefficient. I did this for the NBA and WNBA and then divided the WNBA by the NBA to create scalers for the WNBA against the known NBA PER. These then were scaled against the NBA PER coefficients. They are as follows:
## (Intercept) FG FGA FT FTA ORB
## 0.3289193 53.3482374 29.2755340 15.2740033 25.3355723 47.7225562
## DRB AST STL BLK TOV PF
## 10.9533116 31.6142546 50.7237264 50.6815507 45.8770895 61.2348486
## PTS
## 41.1000000