Exploring the pitcher pace (time between pitches) for games played in a week of the 2015 season

pacestudy.R

library(pitchRx)
library(dplyr)
library(ggplot2)

dat <- scrape(start = "2015-09-05", end = "2015-09-11")
pitches <- inner_join(select(dat$atbat, 
                   batter_name, pitcher_name, inning, 
                   gameday_link, num, url),
                select(dat$pitch,
                   start_speed, pitch_type, sv_id, num, url),
                by = c("num", "url"))

pitches <- mutate(pitches,
                  hours=as.numeric(substr(sv_id, 8, 9)),
                  minutes=as.numeric(substr(sv_id, 10, 11)),
                  seconds=as.numeric(substr(sv_id, 12, 13)),
                  time=3600 * hours + 60 * minutes + seconds)               

get_data <- function(pitcher){
  pdata <- filter(pitches, pitcher_name==pitcher)
  pdata <- mutate(pdata, pa_id = paste(gameday_link, num))
  time_differences <- function(N) diff(filter(pdata, pa_id==N)$time)
  time_to_pitch <- unlist(sapply(unique(pdata$pa_id), time_differences))
  data.frame(Pitcher=pitcher, Time=time_to_pitch)
}

pd <- rbind(get_data("Justin Verlander"),
            get_data("Joe Kelly"),
            get_data("Jon Lester"),
            get_data("Stephen Strasburg"),
            get_data("Bartolo Colon"),
            get_data("Wade Miley"),
            get_data("Max Scherzer"),
            get_data("Jake Arrieta"))
S <- summarize(group_by(pd, Pitcher), M=median(Time))

ggplot(pd, aes(Time)) +
  geom_histogram() +
  facet_wrap(~ Pitcher, ncol=2) +
  geom_vline(xintercept = 20, color="red") +
  xlim(0, 60) +
  geom_vline(data=S, aes(xintercept=M), color="green") +
  xlab("Pace (Seconds)") +
  ggtitle("Pace Data for Eight 2015 Pitchers: Median (Green), 20 (Red)")

PA <- filter(summarize(group_by(pitches, 
                      pitcher_name, gameday_link, num),
                M=median(diff(time)),
                MAD=mad(diff(time)),
                n=n()), is.na(M)==FALSE)

Pitchers <- summarize(group_by(PA, pitcher_name),
                M=sum(M * n) / sum(n),
                N=sum(n))

Pitchers100 <- filter(Pitchers, N > 99)

ggplot(filter(Pitchers100, pitcher_name != "NA"),
       aes(reorder(pitcher_name, M), M)) + 
  geom_point() +
  coord_flip() +
  xlab("Median Time Between Pitches") +
  ylab("Pitcher") +
  ggtitle("Median Pitcher Pace for Pitchers in a Week in 2015 Season") +
  geom_hline(yintercept = 20, color="red")