Graph runs per player with share of team's hits

hit_share.r

library(ggplot2)
library(plyr)
library(reshape)


# Batting.csv from http://baseball1.com/files/database/lahman58-csv.zip

bb <- read.csv("Batting.csv")

bb <-bb[order(bb$yearID, bb$teamID, bb$R),]


bb <- ddply(bb, c('teamID','yearID'), transform,  team_hits_total = sum(H), team_hits_below=cumsum(H)-H)

bb$hit_percentile_floor <- bb$team_hits_below / bb$team_hits_total
bb$hit_percentile_roof<- (bb$team_hits_below + bb$H ) / bb$team_hits_total



# Sample graphs of a single year
# Weight class
    p <- ggplot(subset(bb, yearID == 2005), aes(R, hit_percentile_floor)) 
    p + geom_segment(arrow=arrow(length=unit(0.1,"cm")), aes(yend=hit_percentile_roof,xend=R))  + scale_y_continuous('Player Percentile Rank and Share of Team Hits',formatter = "percent")+ scale_x_continuous('Runs') + facet_wrap(~teamID)