Create a new column that only shows the winning team

问题

I am trying to create a new column that only shows the winning team.

Here is some sample data:

results <- data.frame(
  home_team = c("Scotland", "England", "Scotland", "England", "Scotland", "Scotland",
                "England", "Wales", "Scotland", "Scotland", "England"),
  away_team = c("England", "Scotland", "England", "Scotland", "England", "Wales",
                "Scotland", "Scotland", "England", "Wales", "Wales"),
  home_score = c(0, 4, 2, 2, 3, 4, 1, 0, 7, 9, 2),
  away_score = c(0, 2, 1, 2, 0, 0, 3, 2, 2, 0, 1),
  stringsAsFactors = FALSE
)

this is my code so far:

results <- intl.football.results
first6home <- head(results$home_team)
first6away <- head(results$away_team)
homescore <- (results$home_score)
awayscore <- (results$away_score)

data.frame('winning_team' = 0, results)

for (i in 1:length(results)){
  if(homescore[i] > awayscore[i]){
    homewins <- print("home wins")
  }else if(homescore[i] == awayscore[i]){
    draw <- print("draw")
  }else{
    awaywins <- print("away team wins")
  }
}

I am thinking that I need to somehow rectify the "homewins" to the home_team. The best way I can think of this is by finding the row number of the "homewins" then selecting the rows that the home_team is in. But how do I do this if the data.frame has 30,000+ rows? Sorry this may sound basic but I'm trying!

Thank you everyone for the responses, I will definitely practice them. One last thing, what if I wanted to print out the winning country column and not "home, away, or draw"?

回答1:

The case_when function in dplyr might be a good way to solve this. It seems pretty close to what you're trying to do above, so hopefully its quite intuitive.

Documentation and more examples: https://dplyr.tidyverse.org/reference/case_when.html

I'm passing the name of the the winning team from the corresponding row as the action to take in the case_when, but you can pass in a character string, e.g. 'Home Win', as I've done for the drawn games, if thats the outcome you want.

library(tidyverse)


d <- tibble(
        home_team = c('Scotland', 'England', 'Scotland', 'England', 
                'Scotland', 'Scotland', 'England', 'Wales'), 
        away_team = c('England', 'Scotland', 'England', 'Scotland', 
                'England', 'Wales', 'Scotland', 'Scotland'), 
        home_score = c(0, 4, 2, 2, 3, 4, 1, 0), 
        away_score = c(0, 2, 1, 2, 0, 0, 3, 2))

d %>% 
        mutate(winner = case_when(
                home_score > away_score ~ home_team, 
                away_score > home_score ~ away_team, 
                away_score == home_score ~ 'Drawn Game'))

回答2:

One solution could be to use the data.table package to handle your data. Using this package, the solution to your problem would be (assuming a tie would result in "T", away win in a "A" and home win in a "H")

library(data.table)

setDT(results)
results[
    , w_team := "T"][
    home_score > away_score, w_team := "H"][
    home_score < away_score, w_team := "A"]

回答3:

Base R:

results$who_wins <- with(results,
  ifelse(home_score > away_score, "home wins",
         ifelse(home_score < away_score, "away wins", "draw")))
results
#    home_team away_team home_score away_score  who_wins
# 1   Scotland   England          0          0      draw
# 2    England  Scotland          4          2 home wins
# 3   Scotland   England          2          1 home wins
# 4    England  Scotland          2          2      draw
# 5   Scotland   England          3          0 home wins
# 6   Scotland     Wales          4          0 home wins
# 7    England  Scotland          1          3 away wins
# 8      Wales  Scotland          0          2 away wins
# 9   Scotland   England          7          2 home wins
# 10  Scotland     Wales          9          0 home wins
# 11   England     Wales          2          1 home wins

回答4:

Just for fun, you could also do this by calculating the sign of the score difference and then matching to a lookup vector

lookup <- c('home' = 1, 'away' = -1, 'draw' = 0)
results$winner <- 
  with(results, names(lookup)[match(sign(home_score - away_score), lookup)])

results
#    home_team away_team home_score away_score winner
# 1   Scotland   England          0          0   draw
# 2    England  Scotland          4          2   home
# 3   Scotland   England          2          1   home
# 4    England  Scotland          2          2   draw
# 5   Scotland   England          3          0   home
# 6   Scotland     Wales          4          0   home
# 7    England  Scotland          1          3   away
# 8      Wales  Scotland          0          2   away
# 9   Scotland   England          7          2   home
# 10  Scotland     Wales          9          0   home
# 11   England     Wales          2          1   home

来源：https://stackoverflow.com/questions/58013908/create-a-new-column-that-only-shows-the-winning-team

标签

rstudio

tidy