wanderer | A small roguelike game written in JavaScript | Game Engine library

.iloc[] takes your current row and operates on only that row. So for each row, we first check, whether yout current team is home or away and the we check accordingly if in that row the current team has more goals than the other team, if yes, we set the value for only that row to True:

Use zip to iterate over multiple objects at once instead of nested loops. You will get a tuple of (point, team). Also, eliminate the loop counter variable n by using enumerate. This makes your code more pythonic. Check out the corrected code below:

What you can use for these kind of analyses is the .cumsum() method. If you have floats or ints in a column, you can easily add up the wins up to that point. Make sure to have the order correct (do some simple tests). For example:

Firstly convert your Datetime column into datetime[ns] dtype by using to_datetime() method(If it's already in datetime dtype then ignore/skip this step):

It looks like the column is not missing (na) but has as value '' (empty string). In that case you can do:

I can't test it without the dataframe but I think I see where it goes wrong. In the groupby, you put a list of column names normally. You put 'FT' == 'H' which checks if FT is the same as H (surprisingly it is not).

You might want to groupby lke this:

Perhaps this is what you are looking for?

Improved Your code a bit:

You can use for-loop to iterate through your fixposts & posts then check if the value match if yes increase count value and store in new array .

Demo code :