Similtaneous resignation
06 Jun '05 12:552 editsWith reference to both the "resign when not my turn" thread and the "leaving RHP" thread:
Issue 1) when resigning more than one game at one time, there is a ratings impact on the order of this process. Some oppenents benefit more than others.
Issue 2) If leaving the site, you can't resign all your current games until it is your turn. As one can only resign a few games, the others are left hanging with inevitable TOs, delaying tourneys and sucking up games from non-subscribers' game limit.
There are probably quite a few other issues too but those are enough to get a discussion going.
Solution? Consider "similtaneous resignation" of selected games from the 'My Games' page (and/or a "resign all" button somewhere).
What effect on ratings? Tricky: no answer comes easily to mind. Maybe determine an estimated rating drop, then use an average of this and their current rating to generate a fixed rating for assigning the impact on others. With obvious limits (min level 1000 or so, or something), with a maximum number of two ratings changes for any given opponent. Whatever method needs to be resiliant to those who will inevitably use another ratings-influencing function as a means to inflate their own. Such is life.
The begining of an idea. Shoot away if you so desire.
Issue 1) when resigning more than one game at one time, there is a ratings impact on the order of this process. Some oppenents benefit more than others.
Issue 2) If leaving the site, you can't resign all your current games until it is your turn. As one can only resign a few games, the others are left hanging with inevitable TOs, delaying tourneys and sucking up games from non-subscribers' game limit.
There are probably quite a few other issues too but those are enough to get a discussion going.
Solution? Consider "similtaneous resignation" of selected games from the 'My Games' page (and/or a "resign all" button somewhere).
What effect on ratings? Tricky: no answer comes easily to mind. Maybe determine an estimated rating drop, then use an average of this and their current rating to generate a fixed rating for assigning the impact on others. With obvious limits (min level 1000 or so, or something), with a maximum number of two ratings changes for any given opponent. Whatever method needs to be resiliant to those who will inevitably use another ratings-influencing function as a means to inflate their own. Such is life.
The begining of an idea. Shoot away if you so desire.
