-
We have an objective system for discounting votes where people appear to be gaming the system. This was explained in post #37. It is:
"Any vote where any team is ranked more than 3 places away from its modal rank will be discounted."
The mode is the ranking each team received most. If someone votes a team more than three places away from its modal rank, then the entire vote of that person is discounted.
This year there were three people who were outliers on this measure (in each case only for one team). But I looked at rankings including and excluding these three votes, and the rankings were identical. So the three outlier votes washed out in the overall average, and I thus only presented ranking including all votes (i.e. I did not exclude any votes).
I think this is a sound method for dealing with gaming.
I am happy to share the Excel file if anyone wants to have a look and play around with the data.
You are reading a single comment by @Chuckles and its replies.
Click here to read the full conversation.
Chuckles
really stoked for this and how the playoff has been sorted. seems the only teams not moaning about the split are the ones who actually are involved in the playoff and who really should be the ones maybe disgruntled.
my other point is all the count and points system really doesn't work because it's not very hard for people to heavily low rank people that they think will be around them. therefore giving teams more points than they should. it's sneaky tactics but in the end people aren't exactly honest 24/7
I wouldn't mind seeing the ranking done on a previous league or tournament results. as this would actually give a solid representation of most teams in the league especially when there are mix matches tournament teams! would only be difficult for new players but I think there would be a very small exception that a team of new players or 2/3rds new players would be good enough to play in a top division