R语言if_else/ifelse函数异常：无法正确返回NFL获胜球队名称

Hey there! Let's work through why your attempts aren't behaving as expected and get that winning team column working correctly.

First, Let's Diagnose the Root Issue #

Looking at your output, the tm_won1 and tm_won2 columns show numbers that look like score differences—but they're actually integer codes for factor levels. When you read in your data, R probably converted home_team and away_team to factor variables by default. Functions like ifelse() and case_when() handle factors by returning their underlying integer codes instead of the team names, which is why you're seeing those numbers instead of team labels.

Additionally, if_else() (from dplyr) is strict about type consistency—if the factor levels don't match perfectly between home_team and away_team, it'll return NA for cases where the false condition triggers (like the PHI vs WAS game in your data).

Let's Fix Each Attempt & Get a Working Solution #

1. Fix the Factor Problem First

Before anything else, convert your team columns to character strings to avoid factor-related issues:

# Base R approach
wk1$home_team <- as.character(wk1$home_team)
wk1$away_team <- as.character(wk1$away_team)

# Or dplyr approach (cleaner for multiple columns)
wk1 <- wk1 %>%
  mutate(across(c(home_team, away_team), as.character))

2. Working if_else() Version

Now that teams are stored as characters, if_else() will work as intended (it enforces type consistency, which is good practice):

wk1$tm_won <- if_else(wk1$home_score < wk1$away_score, wk1$away_team, wk1$home_team)

This returns away_team when they score higher, home_team otherwise. For ties, it'll default to home_team—we'll adjust that next if needed.

3. Working ifelse() Version

Base R's ifelse() will now return character strings instead of factor codes:

wk1$tm_won1 <- ifelse(wk1$home_score < wk1$away_score, wk1$away_team, wk1$home_team)

4. Working dplyr + ifelse() with Tie Handling

Your third attempt failed due to factors. With character columns, this will work, and we'll explicitly handle ties:

wk1 <- wk1 %>%
  mutate(tm_won2 = ifelse(home_score < away_score, away_team,
                          ifelse(home_score > away_score, home_team, NA)))

The NA will mark tie games (though NFL regular season ties are rare!).

5. Working case_when() Version

Your fourth attempt had a typo (a instead of NA for ties) and factor issues. Fix that with character columns:

wk1 <- wk1 %>%
  mutate(tm_won = case_when(
    home_score < away_score ~ away_team,
    home_score > away_score ~ home_team,
    TRUE ~ NA_character_  # Use NA_character_ to match character column type
  ))

Using NA_character_ ensures the return type matches the team name strings, avoiding type mismatches.

Example Output After Fixes #

After running the above, your dataset should look like this:

Quick Tip for Future Data Reads #

To avoid factors altogether when reading data, use stringsAsFactors = FALSE in base R's read.csv(), or specify character columns with col_types in read_csv() (from the readr package).

内容的提问来源于stack exchange，提问作者new_r