switching every 2nd row value in pandas - pandas

Question

My dataframe is a list of football games with varying stats, around 300 entries.
game_id team opp_team avg_marks
0 2919 STK BL 122
1 2919 BL STK 114
2 2920 RICH SYD 135
3 2920 SYD RICH 108
I would like to add the opposition stats as a new column for each entry. Resultant dataframe would appear like this
game_id team opp_team avg_marks opp_avg_marks
0 2919 STK BL 122 114
1 2919 BL STK 114 122
2 2920 RICH SYD 135 108
3 2920 SYD RICH 108 135
Any suggestions would be most welcome, I'm new to this forum. I have tried mapping but the entry is conditional on 2 columns, game_id and opp_team.
Ideally I would add it in original spreadsheet but I created a cumulative average for the season in pandas so was hoping there would be a way to incorporate this as well.

Answer 1

You could group on game_id and reverse the avg_marks values
In [725]: df.groupby('game_id')['avg_marks'].transform(lambda x: x[::-1])
Out[725]:
0 114
1 122
2 108
3 135
Name: avg_marks, dtype: int64
In [726]: df['opp_avg_marks'] = (df.groupby('game_id')['avg_marks']
.transform(lambda x: x[::-1]))
In [727]: df
Out[727]:
game_id team opp_team avg_marks opp_avg_marks
0 2919 STK BL 122 114
1 2919 BL STK 114 122
2 2920 RICH SYD 135 108
3 2920 SYD RICH 108 135
Or, get dict mapping from team and avg_marks, then use map on opp_team
In [729]: df['opp_team'].map(df.set_index('team')['avg_marks'].to_dict())
Out[729]:
0 114
1 122
2 108
3 135
Name: opp_team, dtype: int64