R function that provides new variables indicating if plate appearance goes through each possible count.

setup_work.R

setup_work <- function(d2015){
  # d2015 is a Retrosheet play by play dataset
  # removes all non-pitches from PITCH_SEQ_TX
  d2015$pseq <- gsub("[.>123N+*]", "", d2015$PITCH_SEQ_TX)

  # create a b and s sequence
  d2015$pseq <- gsub("[BIPV]", "b", d2015$pseq)
  d2015$pseq <- gsub("[CFKLMOQRST]", "s", d2015$pseq)

  one.string3 <- function(ex, count){
    # create a vector of individual outcomes
    ex.v <- unlist(strsplit(ex,""))
    # remove last X from vector
    ex.v <- ex.v[-length(ex.v)]
    # compute cumulative total of balls and strikes
    n.balls <- cumsum(ex.v == "b")
    n.strikes <- pmin(cumsum(ex.v == "s"), 2)
    the_count <- paste(n.balls, n.strikes, sep="-")
    count %in% the_count
  }
  # applies this function to all strings

  d2015$c01 <- lapply(d2015$pseq, one.string3, "0-1")
  d2015$c10 <- lapply(d2015$pseq, one.string3, "1-0")
  d2015$c02 <- lapply(d2015$pseq, one.string3, "0-2")
  d2015$c11 <- lapply(d2015$pseq, one.string3, "1-1")
  d2015$c20 <- lapply(d2015$pseq, one.string3, "2-0")
  d2015$c12 <- lapply(d2015$pseq, one.string3, "1-2")
  d2015$c21 <- lapply(d2015$pseq, one.string3, "2-1")
  d2015$c30 <- lapply(d2015$pseq, one.string3, "3-0")
  d2015$c22 <- lapply(d2015$pseq, one.string3, "2-2")
  d2015$c31 <- lapply(d2015$pseq, one.string3, "3-1")
  d2015$c32 <- lapply(d2015$pseq, one.string3, "3-2")
  d2015
}