METs for equal but weak players?
Posted By: Timothy Chow
Date: Tuesday, 8 April 2014, at 3:55 p.m.
Typically, people assume that the standard bot MET is the right MET to use for "equal opponents." I'm wondering if anyone has tried to validate this assumption by generating METs from rollouts of (say) GNU 0-ply vs GNU 0-ply, or "X versus X" where X is a bot with some fixed amount of noise added. Though crippling a bot is not a very convincing surrogate for a weak human player, such an experiment would at least give us some information about how robust the standard MET is to perturbing both sides equally.
Messages In This Thread
BGonline.org Forums is maintained by Stick with WebBBS 5.12.