Changing DataFrame columns and data based on logic

Question

So I have a pandas data frame consisting of data from boxing matches and their odds of winning. It's columns are:

[Red_fighter, Blue_fighter, Red_odds, Blue_odds, winner]

I want to change it so that if for example, blue's odds are 'less' than red's that blue gets added to column 'Favourite' and red gets added to 'Underdog' both of which replace 'Red_fighter' and 'Blue fighter'

[favourite, underdog, favourite_odds, underdog_odds, winner]

So if I have:

{'Red_fighter' : 'Tom Jones', 'Blue_fighter' : 'Jack Jackson', 'Red_odds' : 200, 'Blue_odds' : -200 , 'Winner' : 'Blue'}

It becomes:

{'Underdog' : 'Tom Jones', 'Favourite' : 'Jack Jackson', 'Red_odds' : 200, 'Blue_odds' : -200 , 'Winner' : 'Favourite'}

I appreciate any help you can give, I'm a newbie to pandas and data analytics in general, thanks!

Philip Ciunkiewicz · Accepted Answer · 2020-11-08 20:22:49Z

3

You can achieve this using the pd.Series.where method:

df['Underdog'] = df.Red_fighter.where(df.Red_odds < df.Blue_odds, df.Blue_fighter)
df['Favourite'] = df.Red_fighter.where(df.Red_odds > df.Blue_odds, df.Blue_fighter)

df['Underdog_odds'] = df.Red_odds.where(df.Red_odds < df.Blue_odds, df.Blue_odds)
df['Favourite_odds'] = df.Red_odds.where(df.Red_odds > df.Blue_odds, df.Blue_odds)

This method works by replacing the values where a condition is NOT satisfied with values from another series. The remaining values which satisfy the condition are left untouched.

So for example if we have df.A.where(cond, df.B), all rows where cond is True will have values from A and all rows where cond is False will have values from B. There is more information in the documentation.

answered Nov 8, 2020 at 20:22

Philip Ciunkiewicz

2,8183 gold badges15 silver badges25 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Luke Over a year ago

Also, would this make the 'Winner' column automatically have the data of 'Underdog' or 'Favourite' instead of Red or Blue?

Philip Ciunkiewicz Over a year ago

No but you could make it in the same manner: df.Winner = df.Red_fighter.where(df.Winner == 'Red', df.Blue_fighter). Also if the answer helped your question please mark it as accepted to conclude the thread :)

Collectives™ on Stack Overflow

Changing DataFrame columns and data based on logic

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related