How to return value from "other" row using index column in Pandas

Question

My data frame df is comprised of pairs of rows with matching index columns ['game_id'] and differing values otherwise. If I want to create a column that references a value from the other row of that same index, how do I do that?

For example, if I'm in the first row of ['game_id'] and I want to create a column for ['FGA'] values of the pair's match, how do I do that? The first row would return 85 while the second would return 90 . In the ['game_id'] == 21900017 rows , the first row would return 78 while the second row would return 109 .

game_id   Team  PTS FGM Miss FTA TOV Foul ORB DRB  UA  FGA    FG%   A%
21900008  BOS   93  33  57   34  10   29  10   31  15   90   0.367  0.545
21900008  PHI   107 37  48   36  15   34  12   50  13   85   0.435  0.649
21900017  BOS   112 42  67   16  9    21  21   35  21   109  0.385  0.5
21900017  TOR   106 37  41   17  23   28  5    41  15   78   0.474  0.595

Answer 1

You can use groupby() and apply() and reverse the index of the grouped series to find the value of the pair:

df['FGA_pair'] = df.groupby('game_id')['FGA'].apply(lambda x: x.iloc[::-1]).reset_index(drop=True)

Yields:

    game_id Team  PTS  FGM  Miss  FTA  ...  DRB  UA  FGA    FG%     A%  FGA_pair
0  21900008  BOS   93   33    57   34  ...   31  15   90  0.367  0.545        85
1  21900008  PHI  107   37    48   36  ...   50  13   85  0.435  0.649        90
2  21900017  BOS  112   42    67   16  ...   35  21  109  0.385  0.500        78
3  21900017  TOR  106   37    41   17  ...   41  15   78  0.474  0.595       109