How to filter a pandas DataFrame to show the largest value for a given column?

Question

I have NBA data for every game played until January of 2023. I went ahead and combined all the points and now have it as this:

I wanted to filter the dataframe to show the most points scored by a player for a given season. For example:

Season	Name	Pts
2022	Player 1	2938
2021	Player 2	2111

most_points = nba_data.groupby(by=['Season', 'Name'])[['Pts']].sum()
most_points.sort_values(by='Pts', ascending=False)

Reset_index then sort by points and drop_duplicates by name. — Scott Boston, Feb 26 '23 at 03:50
@Scott `groupby.idxmax` should always be preferred, sorting is `O(n*logn)` while `groupby.idxmax` is `O(n)` ;) — mozway, Feb 26 '23 at 06:39

Jamiu S. · Answer 1 · 2023-02-26T07:54:06.627

0

Using pandas.DataFrame.loc and df.groupby().idxmax() in the below approach will produce your expected output

most_pts = nba_data.loc[nba_data.groupby(['Season'])['Pts'].idxmax()]
print(most_pts)

edited Feb 26 '23 at 07:54

answered Feb 26 '23 at 05:12

Jamiu S.

1 Answers1