A variable that indicates who win

Question

I have a database where four of its columns are: score_home, score_away, home_id and away_id.

I expect to get a variable whose rows contain the winning ID in each game.

Index	gRes
0	GB
1	GB

For that, I tried with the following code

team_f['gRes'] = 0
if team_f['score_home'] > team_f['score_away']:
    team_f['gRes'] = team_f['home_id']
else:
    team_f['gRes'] = team_f['away_id']

and i get the following error

The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

Could you suggest how to correct the error or, failing that, any alternatives to build the variable?

what about tied games with level score? – Naveed Oct 26 '22 at 16:43 — Naveed, Oct 26 '22 at 16:43

score 0 · Answer 1 · answered Oct 26 '22 at 16:40

This is ambiguous because of how you compare the two columns.

Try np.where

team_f['gRes'] = np.where(team_f['score_home'] > team_f['score_away'], team_f["home_id"], team_f["away_id"])

Please include a minimal example of your data if that does not work.

Edit: If draws are possible, np.select is preferred. (See Naveed's answer)

score 0 · Accepted Answer · answered Oct 26 '22 at 16:40

what if the score are tied? you'll have three conditions, and np.Select helps

# define a list of conditions, and corresponding list of values, 
# when condition is true, 
# and default value as a third attribute

df['gRes'] = np.select([team_f['score_home'] > team_f['score_away'], #cond 1
                       team_f['score_home'] < team_f['score_away']   #cond 2
                       ],
                       [team_f['home_id'],  # cond1 true
                        team_f['away_id']], # cond2 true
                       'drawn'              # default, game tied
                      )

PS: if you share a data(preferably as a code) or text, I'd be able to share the result

A variable that indicates who win

2 Answers2