Given a data frame like this one :
player draftyear draftpick season team games games_start mins points
Luke Babbitt 2010 16.0 2010 POR 24 0 137 36
Luke Babbitt 2010 16.0 2011 POR 40 4 537 202
Luke Babbitt 2010 16.0 2012 POR 62 0 730 244
Luke Babbitt 2010 16.0 2013 NOP 27 2 473 170
Luke Babbitt 2010 16.0 2014 NOP 63 19 830 256
Luke Babbitt 2010 16.0 2015 NOP 47 13 845 327
Luke Babbitt 2010 16.0 2016 MIA 68 55 1065 324
Luke Babbitt 2010 16.0 2017 ATL 37 9 570 226
Luke Babbitt 2010 16.0 2017 MIA 13 5 145 33
We see that Luke Babbitt played for two teams in the 2017 season. I would like to condense the stats in the season where he played on multiple teams.
The way I would like to condense it would preserve player, draftyear, draftpick, season. I would then like to make a list out of the teams he played for, and then add together games, games_Start, mins, and points.
The output should look something like this:
player draftyear draftpick season team games games_start mins points
Luke Babbitt 2010 16.0 2010 POR 24 0 137 36
Luke Babbitt 2010 16.0 2011 POR 40 4 537 202
Luke Babbitt 2010 16.0 2012 POR 62 0 730 244
Luke Babbitt 2010 16.0 2013 NOP 27 2 473 170
Luke Babbitt 2010 16.0 2014 NOP 63 19 830 256
Luke Babbitt 2010 16.0 2015 NOP 47 13 845 327
Luke Babbitt 2010 16.0 2016 MIA 68 55 1065 324
Luke Babbitt 2010 16.0 2017 ['ATL', 'MIA'] 50 14 715 259
So far the only thing I have is:
df.groupby(['player', 'season'], as_index=False)[['games', 'games_start', 'mins','points']].agg(sum)
However I am having a hard time figuring out how to do the other two functions in the same line, where I keep certain entries the same and for the other ones I turn it into a list.
