So to get this you first need to get a calc field that gets the win %:
SUM(IF [FTR] = 'AWins' OR [FTR] = 'Hwins' THEN 1 END)/COUNTD([Game ID])
This can then be used to rank the referees:

Now the reason that it may not be working for you with your technique is that you're sorting on COUNTD(Wins) which is the total number of wins, not the percentage wins for the ref. So someone that has just played more games may come up higher in the rank
Now you have the calc field, you can go back to your report and sort on the new field:

I rearranged the legend so you can see that the ref with the best % wins are shown first (red and blue bars)
If you don't want it sorted by win %, then change the calc field to:
SUM(IF [FTR] = 'AWins' OR [FTR] = 'Hwins' THEN 1 END)
For the COUNTD of games, if you only have the date and the game available and want to create an ID from that that is unique, create a calc field like this:
game-date-id = STR([game]) + STR(' ') + STR(date)
This will then be used in your COUNTD if statement:
SUM(IF [FTR] = 'AWins' OR [FTR] = 'Hwins' THEN 1 END)/COUNTD([game-date-id])