Plots trajectories of strikeout rates, home run rates, and hit-in-play rates for players with similar batting averages

plot_career_trajectory_rates.R

# requires packages
# dplyr, Lahman, ggplot2

# some preliminary work

library(dplyr)
library(Lahman)

get.birthyear <- function(player.id){
  birthdata <- select(filter(Master, playerID==player.id),
                      birthMonth, birthYear)
  with(birthdata, 
       ifelse(birthMonth >= 7, birthYear + 1, birthYear))
}
Batting1 <- summarize(group_by(Batting, playerID, yearID),
                      AB=sum(AB),
                      H=sum(H),
                      SO=sum(SO),
                      HR=sum(HR))
Master1 <- mutate(Master,
                   BirthYear = get.birthyear(playerID))
Batting1 <- mutate(inner_join(Batting1, 
                      select(Master1, playerID, BirthYear),
                      by="playerID"), Age=yearID - BirthYear)
S <- summarize(group_by(Batting1, playerID),
                      MinYear = min(yearID),
                      MaxYear = max(yearID),
                      MidCareer = (MinYear + MaxYear) / 2,
                      CareerAB = sum(AB),
                      CareerH = sum(H),
                      CareerAVG = round(CareerH / CareerAB, 3))
Batting1 <- inner_join(Batting1, 
                select(S, playerID, MidCareer, CareerAB, CareerAVG),
                      by="playerID")

# look at players with similar career batting averages with a min
# number of career AB in a particular
# era -- look at trajectories of SO Rate = SO / AB and
# HIP Rate = H / (AB - SO)
# idea is that the SO Rates should be more stable over time than
# HIP rates

# by default find players with at least 3000 AB and career AVG of .300 
# target midcareer year that is input
# allow errors of 4 for midcareer and .002 for target career AVG

compare_rates <- function(target.year, Career_eps = 4,
                          AVG_target = .300, 
                          AVG_eps = .002,
                          Career_target = 3000){
require(ggplot2)
S_select <- filter(S,
              abs(MidCareer - target.year) <= Career_eps,
             abs(CareerAVG - AVG_target) <= AVG_eps,
             CareerAB >= Career_target)

Batting_select <- filter(Batting1,
                         playerID %in% S_select$playerID)
Batting_select <- inner_join(Batting_select,
                        select(Master, playerID, nameFirst, nameLast),
                        by="playerID")
Batting_select <- mutate(Batting_select,
                         Name=paste(nameFirst, nameLast))

if(dim(Batting_select)[1] == 0) cat("No players matched the criteria.")

if(dim(Batting_select)[1] > 0) {
plot1 <- ggplot(Batting_select, aes(Age, SO / AB)) +
  geom_point(color="red") + geom_smooth(se=FALSE) + 
  ggtitle("Strikeout Rates") +
  facet_wrap(~ Name, ncol=2)

plot2 <- ggplot(Batting_select, aes(Age, HR / (AB - SO))) +
  geom_point(color="red") + geom_smooth(se=FALSE) + 
  ggtitle("Home Run Rates") +
  facet_wrap(~ Name, ncol=2)

plot3 <- ggplot(Batting_select, aes(Age, (H - HR) / (AB - SO - HR))) +
  geom_point(color="red") + geom_smooth(se=FALSE) + 
  ggtitle("Hit-In-Play Rates") +
  facet_wrap(~ Name, ncol=2)
print(plot1)
print(plot2)
print(plot3)
}
}