I think skill is determined different from player to player. With that being said there's no one measurement that can determine skill alone. There's a load of solo pvpers that if they stayed partied up their stats would look totally different. And the same goes for people who stay partied up. They might appear "average" solo. Then, you have the players who when they first start, use that 1st character as a test. A means to get good. That player could take all that knowledge gained from practice and apply it to character 2 and 3. But still end up with an average kd because of how much time was spent practicing on character 1. What I'm trying to say is you really can't judge using the general methods. You'd have to first know a little bit about the player to eliminate plenty of variables that are not generally taken into account.