Tiamat, the numbers are only a starting point. Its done in this way to get the data out and create a larger set of data for statistical purposes. I will say i'd rather keep it to discreet integer numbers, so no fractions. If it comes to that, like I said, i'd like to see the 2 mods just hash it out and the more experienced mod's score goes. After the initial set of data is done, there will of course be the following issues:
1) not all matchups done
2) errors due to lack of knowledge and experience
To address both issues, I would appoint 2-3 persons to head up a Ranking project discussion. They would be responsible for doing a thorough review of each character by starting a thread for each character and walking though the matchups. I'd let each discussion last a week and then decide if the matchup numbers would change. The committee could change, but i'd like a list of all possible committee members posted, but only 2-3 would need to participate at a time. Final scores would would be decided by the committee members, but any player would be free to give input. This addresses the concern about exp AND breadth of input.
Ideally I'd like to see a review happen once every couple of weeks. So that new information can be constantly incorporated. I do agree that EVO and nats will be useful, but I don't see how at least getting an initial matchup chart will not be useful. We can compare the initial data against the tourney results and see how it all pans out.
So for example
Ranking project team:
Aris, Johnny B, Fhwoarang, Tiamat, STD, Ceirnian, KDZ, OOF, Saitoh, and Hayate
So for example: week 1: Maxi - Initial Ranking project team members: Johnny, Tiamat, and Aris ( names out of my ass )
at the end of week 1: Maxi vs Sets, NM, Seig are changed. Rest are ok.
Week 3: Taki - New member: STD - Johnny out
at end of week 3: Taki vs Mitsu, Cass, Sophie, Ast changed. Rest are ok.
etc etc etc until all chars are reviewed.
