calculate the spin_rate distribution under condition : HIT or STRIKE

spin_rate_vs_hit_or_strike.r

library(dplyr)
library(magrittr)
library(data.table)
library(ggplot2)

# pitch f/x data 2013
dat = fread("2013.csv")
dat = dat %>% filter(sv_id !="NA")

# hit event
hits = c("In play, run(s)", "In play, no out")
strikes = c("Called Strike", "Swinging Strike")

## data processing
# hit flag
dat = dat %>% mutate(h_fl = ifelse(des %in% hits, "HIT", "NOT")) %>% mutate(h_fl = as.factor(h_fl))
# strike flag
dat = dat %>% mutate(strike_fl = ifelse(des %in% strikes, "STRIKE", "NOT")) %>% mutate(strike_fl = as.factor(strike_fl))
# hit or strike
dat = dat %>% filter(h_fl == "HIT" | strike_fl =="STRIKE")
dat = dat %>% mutate(hit_or_strike = ifelse(h_fl=="HIT", "HIT", "STRIKE"))

# visualize
pingr(
dat %>% filter(sv_id!="NA") %>% select(spin_rate, hit_or_strike) %>% 
  mutate(spin_rate = as.numeric(spin_rate)) %>%
  ggplot() + geom_histogram(aes(x=spin_rate, col="red"), stat="density", binwidth=100) + facet_grid(hit_or_strike ~ .) + 
  ggtitle("spin_rate distribution (Hit or not)") + 
  theme(legend.position="NULL") + theme(plot.title=element_text(face="bold", size=20))
)
ping(3)