|
Let the admin specify in their listing the average number of players they have.
Ratio the votes the game receives against this. If there are more than 110% votes compared to the reported number of players the game has, discard them as erroneous.
If an admin reports their average players as 100, and they receive 100 votes, then they get 100% weight.
If they get 130 (> 110, therefore affected) votes, detract somewhat from the overall weight of the votes.
If they get < 90 votes, then detract the value of the votes.
Thus:
votes weighted % value
100 votes -- 100
130 votes -- 90
90 votes -- 85
Then list based on the % of votes compared to another, as opposed to the raw quantity.
Benefits:
o This will ensure the admins take an active hand in keeping their listings up to date
o Players will not spam-vote their favorite game, because they will actually lose weight if they do
o It will report a more accurate relationship to the quantity of players relative to the quality
Flaws:
o Those same spam-voters can now vote for others, to reduce the overall weight of that listing
o No system is going to report what everyone wants it to report.
Thoughts: A raw-number system such as currently is in place does nothing to show anything other than who is able to generate the most votes in a given period. Yes, one could argue that those with more players are also the better games, but this is not always the case. Some games, such as Armageddon, Southlands, Towers of Jadri, and others, cater to a specific clientelle and would not necessarily flourish with hundreds of players on at a time.
--Bert
|