R/clean_bowling_data.R

In ropenscilabs/cricketdata: International Cricket Data

# Function to clean bowling data.
# Works with career or innings data

clean_bowling_data <- function(x) {
  # Make names easier to interpret
  vars <- colnames(x)
  vars[vars == "Mat"] <- "Matches"
  vars[vars == "Inns"] <- "Innings"
  vars[vars == "Mdns"] <- "Maidens"
  vars[vars == "Wkts"] <- "Wickets"
  vars[vars == "BBI"] <- "BestBowlingInnings"
  vars[vars == "BBM"] <- "BestBowlingMatch"
  vars[vars == "Ave"] <- "Average"
  vars[vars == "Econ"] <- "Economy"
  vars[vars == "SR"] <- "StrikeRate"
  vars[vars == "4"] <- "FourWickets"
  vars[vars == "5"] <- "FiveWickets"
  vars[vars == "10"] <- "TenWickets"
  vars[vars == "Start Date"] <- "Date"
  colnames(x) <- vars

  # Fix classes for all variables
  if ("Maidens" %in% vars) {
    x$Maidens <- as.integer(x$Maidens)
  }
  if ("Balls" %in% vars) {
    x$Balls <- as.integer(x$Balls)
  }
  x$Runs <- as.integer(x$Runs)
  x$Wickets <- as.integer(x$Wickets)
  x$Innings <- as.integer(x$Innings)

  career <- ("Matches" %in% vars)

  if (career) {
    x$Matches <- as.integer(x$Matches)
    if ("Span" %in% vars) {
      x$Start <- as.integer(substr(x$Span, 1, 4))
      x$End <- as.integer(substr(x$Span, 6, 9))
    }
    if ("FourWickets" %in% vars) {
      x$FourWickets <- as.integer(x$FourWickets)
    }
    if ("FiveWickets" %in% vars) {
      x$FiveWickets <- as.integer(x$FiveWickets)
    }
    if ("TenWickets" %in% vars) {
      x$TenWickets <- as.integer(x$TenWickets)
    }
  } else {
    # Add participation column
    if ("Overs" %in% vars) {
      x$Participation <- participation_status(x$Overs)
      x$Overs[x$Participation != "B"] <- NA
    }
    x$Date <- lubridate::dmy(x$Date)
    x$Opposition <- stringr::str_replace_all(x$Opposition, "v | Women| Wmn", "")
    x$Opposition <- rename_countries(x$Opposition)
  }
  if ("Overs" %in% vars) {
    x$Overs <- as.numeric(x$Overs)
  }

  # Recompute average to avoid rounding errors
  if ("Average" %in% vars) {
    x$Average <- x$Runs / x$Wickets
  }

  # Recompute economy rate to avoid rounding errors
  if ("Balls" %in% vars) {
    balls <- x$Balls
  } else {
    balls <- trunc(x$Overs) * 6 + (x$Overs %% 1) * 10
  }
  if ("Economy" %in% vars) {
    x$Economy <- as.numeric(x$Economy)
    ER <- x$Runs / (balls / 6)
    # Don't modify values if they differ by more than 0.05
    differ <- abs(round(ER, 2) - x$Economy) > 0.05
    differ[is.na(differ)] <- FALSE
    x$Economy[!differ] <- ER[!differ]
  }

  # Recompute strike rate
  if ("StrikeRate" %in% vars) {
    x$StrikeRate <- balls / x$Wickets
  }

  # Extract country information if it is present
  # This should only be required when multiple countries are included
  country <- (length(grep("\\(", x[1, 1])) > 0)
  if (country) {
    x$Country <- stringr::str_extract(x$Player, "\\([a-zA-Z \\-extends]+\\)")
    x$Country <- stringr::str_replace_all(x$Country, "\\(|\\)|-W", "")
    x$Country <- rename_countries(x$Country)
    x$Player <- stringr::str_replace(x$Player, " \\([a-zA-Z  \\-]+\\)", "")
  }

  # Re-order and select columns
  vars <- colnames(x)
  if (career) {
    varorder <- c(
      "Player", "Country", "Start", "End", "Matches", "Innings", "Overs", "Balls", "Maidens", "Runs", "Wickets",
      "Average", "Economy", "StrikeRate", "BestBowlingInnings", "BestBowlingMatch", "FourWickets", "FiveWickets", "TenWickets"
    )
  } else {
    varorder <- c(
      "Date", "Player", "Country", "Overs", "Balls", "Maidens", "Runs", "Wickets",
      "Economy", "Innings", "Participation", "Opposition", "Ground"
    )
  }
  varorder <- varorder[varorder %in% vars]

  return(x[, varorder])
}